IDENTIFICATION OF NOVEL CELLULAR COMPONENTS ASSOCIATED WITH HIV-1 EARLY NUCLEOPROTEIN COMPLEXES CAMERON JAY SCHWEITZER A DISSERTATION

Transcription

1

2

3 IDENTIFICATION OF NOVEL CELLULAR COMPONENTS ASSOCIATED WITH HIV-1 EARLY NUCLEOPROTEIN COMPLEXES BY CAMERON JAY SCHWEITZER A DISSERTATION Submitted to the Faculty of the Graduate School of Creighton University in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in the Department of Medical Microbiology and Immunology Omaha, Nebraska May 3, 2012 ii

4 ABSTRACT The human immunodeficiency virus type 1 (HIV-1) is a member of the subfamily lentiviridae and the etiological agent of acquired immune deficiency syndrome (AIDS). Treatment of HIV-1 infection includes a multi-drug regimen that severely limits active replication but is unable to completely eradicate the virus. Current drugs target several viral proteins to inhibit critical functions; however HIV-1 multi-drug resistance continues to rise. Several genomic screens have highlighted the importance of numerous cellular proteins in HIV-1 replication. The viral nucleoprotein complexes (NPCs) formed during early HIV-1 infection represent potential targets for antiviral therapy. Despite extensive research the structure and composition of these structures remain poorly characterized. To better characterize these structures, two methods were developed to purify HIV-1 complexes for proteomic analysis and a potential co-factor was extensively studied. The first purification method used biotin tagging of two HIV-1 proteins, IN and MA. When the HIV-1 BAS-containing proteins IN and MA were expressed in conjunction with the biotin ligase BirA, each protein was efficiently and specifically biotinylated. The MA virus was fully competent for infectivity and replication both with and without biotin. However, the IN virus was approximately 50% less infectious without biotin and the addition of biotin rendered this virus noninfectious. The insertion and addition of biotin severely disrupted integration activity of this virus. HIV-1 IN and MA protein complexes were affinity purified at 4 hours post infection and candidate cellular proteins were identified by mass spectrometry. Several known HIV-1 interacting proteins were identified as well as previously undescribed proteins. The cellular protein leucine rich PPR-motif containing (LRPPRC) was further examined in HIV-1 infection. iii

5 The next study aimed to determine the importance of LRPPRC in HIV-1 replication. During early HIV-1 infection LRPPRC was associated with both viral RNA and DNA. To ascertain whether LRPPRC expression was required for HIV-1 infection, endogenous protein levels were depleted in cells by short hairpin RNA interference. Both transient and stable depletion of LRPPRC significantly reduced HIV-1 infectivity. Furthermore, LRPPRC expression was required for nuclear import and preintegration complex stability, but not reverse transcription. Lastly, LRPPRC expression was not essential for viral release and vrna incorporation. Combined, these data identified LRPPRC as a HIV-1 factor that is critical for HIV-1 replication through two different mechanisms. In the final study, functional HIV-1 RTCs and PICs were partially purified by velocity gradient centrifugation and fractionation, concentrated, trypsin digested, and analyzed by LC-MS/MS. A comparison of the HIV-1 samples to parallel uninfected control samples was used to identify associated cellular factors. A total of seven parallel infected and control biological replicates were completed, producing a data set of 8340 proteins. Several previously discovered HIV-1 factors were identified in our screen including GANAB, CD58, VPS37B, BTRC, and CD28. The database searches identified 99 proteins unique to the viral samples. Identification of two proteins by Western blot and immunoprecipitation confirmed the MS results. These data identify a number of additional candidate proteins that may regulate the early events of HIV-1 replication. iv

6 ACKNOWLEDGEMENTS I would first like to thank my mentor, Dr. Michael Belshan for his superior guidance throughout my graduate education. By his example I learned not only scientific technique, but also scientific integrity which I will take with me while moving forward with my career. I would also like to thank my graduate committee members Dr. Jason Bartz, Dr. Patrick Swanson, Dr, Nancy Hanson, and Dr. Pawel Ciborowski for their direction and thought-provoking questions. I owe many thanks to the entire faculty and staff of the medical microbiology and immunology department for their aid and assistance. Special thanks to Dr. Greg Perry for the flow cytometry analyses and Dr. Ron Cerny with the proteomics core at UNL. Second, I need to thank the many lab members that have passed through the Belshan lab. Meghan Donnellen, Alison Larsen, Mack Savage, Zach Ploeger, Christian Madson, and John Matthews were very supportive early in my graduate work and also provided a fun lab environment. Many of them became more than just co-workers and I am grateful for their friendship. I would also like to thank the rest of my fellow graduate students, many of which were instrumental in my classroom learning and studying. I also need to thank the various members of the Ciborowski lab, such as Teena Jagadish, Nicole Haverland, and Wojciech Rozek for their constant aid with the proteomic studies. Lastly, I would like to thank my friends and family. My mother and father are a source of constant support. I would not be who I am today without their love. I would also like to thank my two sisters for their continuous encouragement. Also I would like to v

7 thank my mother-in-law and father-in-law for the prayers and encouragement throughout this journey. Most of all I need to thank my wife and best friend, Rachel. Thank you for your unending love and steadfast strength in the face of this daunting challenge. I truly could not have done this without you. I want to thank God for his many blessings, especially my unborn son, whom I love more than anything already. vi

8 DISSERTATION ORGANIZATION This dissertation represents the research performed throughout my graduate education. Chapter I is an introduction to HIV-1 and our current knowledge of virus replication. It also addresses the current problems involved in treating HIV-1 infection and states the goals for the research presented. Chapter II describes the development of a biotinylation system to affinity purify HIV-1 protein complexes in vitro. A majority of this work was published in the Journal of Virological Methods (Belshan et al., 2009). The HIV-1 molecular clones containing the biotinylation sequence were provided by Dr. Alan Engelman. Most of the research for this chapter I performed myself with the exception of the 293T cell line expressing BirA which was done by Dr. Michael Belshan. In addition, the processing of the affinity purified samples was performed with the indispensable help of John Matthews. Meghan Donnellan performed the integration reactions. Lastly, the mass spectrometry was performed at the Nebraska Center for Mass Spectrometry with Dr. Ronald Cerny. Chapter III presents research into the role of the cellular protein LRPPRC which was identified using the biotinylation system in HIV-1 replication. This work is in revision for publication in PLoS ONE. The data presented was done by me with the exception of Figure 3.4C which was done by Dr. Michael Belshan, the cell cycle and JC- 1 flow cytometry was performed by Dr. Greg Perry, Meghan Donnellan constructed the ptz-ltr standard, and Christian Madson performed the integration reactions. Chapter IV details research examining the proteome of functional PICs and RTCs isolated by velocity gradient centrifugation. This work is currently in preparation vii

9 for publication in the Journal of Proteome Research. The preliminary work for isolation of PICs and RTCs was performed by Meghan Donnellan and Christian Madson. I did all gradient isolations for mass spectrometry (MS). These studies were performed in collaboration with the laboratory of Dr. Pawel Ciborowski. The processing and MS was performed by Teena Jagadish. Data analysis of the MS results was done by Nicole Haverland and myself. Appendix A includes research examining the cellular protein barrier to autointegration factor (BAF) in relation to HIV-1 and MLV infection. The cells and plasmids for this work were kindly provided by Dr. Matthew Wiebe from the University of Nebraska-Lincoln. I generated all data contained within this section. viii

10 LIST OF TABLES Table 1.1: Currently available antiretroviral drugs approved for HIV-1 treatment... 4 Table 1.2: Cellular proteins important for early HIV-1 infection Table 2.1: HIV-1 proteins identified in the infected samples by MS analysis Table 2.2: Candidate HIV-1 interacting cellular proteins identified by MS Table 4.1. Proteome Discoverer Protein Hit Summary Table 4.2. Selected proteins unique to infected samples Table 4.3. Selected proteins enriched in infected samples Table 4.4. Selected proteins unique to control samples Table 4.5. Selected proteins enriched in control samples Table 4.6. Unique proteins identified in infected fractions LIST OF FIGURES Figure 1.1: The HIV-1 genome... 6 Figure 2.1: Schematic illustrating the BAS-BirA system Figure 2.2: Construction of BAS-containing IN and MA molecular clones Figure 2.3: Specific in vivo biotinylation of IN and MA Figure 2.4: Affinity purification of IN and MA in stable BirA expressing cells.. 36 Figure 2.5: Replication and infectivity profiles of BAS molecular clones Figure 2.6: Biotinylation ablates integration activity Figure 2.7: Biotinylation does not affect reverse transcription Figure 2.8: Purification of biotinylated IN and MA protein complexes ix

11 Figure 2.9: Validation of candidate proteins in affinity purified samples Figure 3.1: LRPPRC associates with HIV-1 RNA and DNA complexes during early infection Figure 3.2: LRPPRC does not associate with IN or MA in co-immunoprecipitations Figure 3.3: Stable knockdown of LRPPRC alters the subcellular localization of several proteins Figure 3.4: Knockdown of LRPPRC impairs HIV-1 infection Figure 3.5: LRPPRC depletion reduces HIV-1 nuclear import and PIC formation Figure 3.6: Knockdown of LRPPRC does not affect viral RNA encapsidation or virus release Figure 3.7: Characterization of shlrpprc cells Figure 4.1: Assessment of infection length and extraction method Figure 4.2: Schematic depiction of the methods used in this analysis Figure 4.3: Isolation of functional HIV-1 preintegration complexes and reverse transcription complexes Figure 4.4: Venn diagrams of proteins identified in the Proteome Discoverer analysis Figure 4.5: XRCC6 is present in the infected samples and is associated with HIV-1 DNA Figure 4.6: Validation of candidate proteins identified by Proteome Discoverer Figure 4.7: Venn diagrams of proteins identified in the ProteoIQ analysis Figure 4.8: The top two scoring pathways identified by ingenuity pathway analysis Figure 4.9: The remaining pathways identified by ingenuity pathway analysis x

12 Figure 4.10: Validation of candidate proteins identified by ProteoIQ Figure A.1: Knockdown of BAF does not decrease MLV infectivity Figure A.2: WT and mutant MAAAQ BAF does not alter HIV-1 infectivity Figure A.3: BAF does not colocalize with HIV-1 during early infection xi

13 LIST OF ABBREVIATIONS AIDS Acquired immune deficiency syndrome ART Antiretroviral therapy BAS Biotin acceptor sequence CA Capsid env Envelope gag Group specific antigen h.p.i. hours post infection HIV-1 Human immunodeficiency virus type 1 HRP Horseradish peroxidase IN Integrase LRPPRC Leucine-rich PPR motif-containing LTR Long terminal repeat MA Matrix MLV Murine leukemia virus NC Nucleocapsid Nef Negative factor xii

14 NLS Nuclear localization sequence NPC Nucleoprotein complex PIC Preintegration complex pol Polymerase PR Protease Rev - Anti-repression transactivator protein RT Reverse transcriptase RTC Reverse transcription complex SA - Streptavidin Tat Transactivating regulatory protein Vif Viral infectivity factor Vpr Viral protein R Vpu Viral protein U xiii

15 TABLE OF CONTENTS List of Tables... ix List of Figures... ix List of Abbreviations... xii Chapter I: Introduction HIV-1 Infection and Treatment Human Immunodeficiency Virus Type The HIV-1 Lifecycle Literature Review of Early HIV-1 Infection The Viral Nucleoprotein Complexes HIV-1 Uncoating HIV-1 Nuclear Import Cellular Proteins Important for Early HIV-1 Infection Research Goals Chapter II: Identification of HIV-1 Interacting Cellular Proteins by Affinity Purification of Integrase and Matrix Background Methods and Materials Plasmids Cell Culture Virus Preparations In Vitro Integration Assays Affinity Purification Mass Spectrometry Results Characterization of BAS molecular clones xiv

16 2.3.2 Identification of Proteins Associated with IN and MA Proteins Complexes in vivo Discussion Future Studies Chapter III: Knockdown of the Cellular Protein LRPPRC Attenuates HIV-1 Infection Background Methods and Materials Plasmids Cell Culture and Virus Preparations T shlrpprc Cells Viral Infectivity Assays Reverse Transcription and Nuclear Import Quantification ptz19r-ltr Integration Assays Cell Cycle and Proliferation Assays Virus Release Assay and Viral RNA Quantification Immunoprecipitation Assays Results Association of LRPPRC with HIV-1 RNA and DNA During Early Infection but not Integrase or Matrix Cell Compartment Specific Depletion of LRPPRC in 293T Cells LRPPRC Knockdown Decreases HIV-1 Infectivity Loss of LRPPRC Reduces HIV-1 PIC Formation and Nuclear Import LRPPRC is Not Required for Virus Assembly and Release xv

17 3.3.6 Growth Characteristics and Expression Profiles of the shlrpprc Cell Lines Discussion Future Studies Chapter IV: Proteomic Analysis of Early HIV-1 Nucleoprotein Complexes Background Materials and Methods Cell Culture and Viral Infections Velocity Gradient Centrifugation and Fractionation Viral DNA Quantification and Integration Assays In-solution Tryptic Digest Protein Identification by Nano-LC-MS/MS Data Parsing and Ingenuity Pathway Analysis Western Blot Immunoprecipitation-PCR Results and Discussion Optimal Infection Parameters Proteome Discoverer Validation of the Extraction Method Validation of Proteome Discoverer Candidate Proteins ProteoIQ Validation of ProteoIQ candidate proteins Future Studies xvi

18 Appendix A: Barrier to Autointegration Factor and Retroviral Infection A.1 Background A.2 Materials and Methods A.2.1 Cell Culture and Infections A.2.2 Luciferase Assays A.2.3 Fluorescent Microscopy A.3 Results A.4 Discussion Chapter V: References xvii

19 CHAPTER I: INTRODUCTION The human immunodeficiency virus type 1 (HIV-1) is the etiological agent of acquired immune deficiency syndrome (AIDS). Currently the United Nations estimates that 33.2 million people are infected worldwide (UNAIDS, 2010). The untreated disease causes a progressive depletion of T lymphocytes within the immune system and allows for a myriad of opportunistic pathogens to attack the infected individual. There is no vaccine available to stop the spread of HIV-1, thus the greatest line of defense against this pandemic is antiretroviral therapy (ART) which is commonly a combination of 3 drugs. Currently, there are 32 drugs approved for the treatment of HIV-1. While these therapies have vastly improved the clinical outcome of patients, but there is no cure and patients must maintain ART indefinitely. This often leads to problems with adherence especially due to the adverse side effects associated with these drugs. The life-long therapy also increases the risk of multi-drug resistant mutants which can lead to treatment failure and progression of disease. These problems must be addressed by identifying new targets for drug therapy to increase the pool of existing drugs and ultimately reduce emerging drug resistance. 1.1 HIV-1 INFECTION AND TREATMENT HIV-1 infection can be divided into two phases, the acute phase and the chronic phase. Infection begins with exposure of the virus to mucosal surfaces or the blood, with more than 80% of infections occurring via the mucosal surfaces and the remaining 20% of infections occur via percutaneous or intravenous exposure (UNAIDS, 2010). Immediately after exposure, HIV-1 begins replication in the mucosa, submucosa, and 1

20 lymphoreticular tissues, although vrna cannot be detected in the blood for approximately 7-21 days (Keele et al., 2008; Lee et al., 2009). The understanding of early acute infection is still poorly understood, but evidence from simian immunodeficiency virus (SIV) in animal models suggest that CD4+ T cells and Langerhans cells are the first targets infected (Boggiano and Littman, 2007; Hladik et al., 2007). Other cell types such as dendritic cells may also be important at this early stage (Lackner and Veazey, 2007), although recent evidence suggests these cells are poor targets when compared to CD4+ T cells (Li et al., 2010; Salazar-Gonzalez et al., 2009). Regardless of the initial cells infected, HIV-1 eventually converges on the lymphoreticular system in the gastrointestinal tract (Brenchley et al., 2004; Mattapallil et al., 2005; Mehandru et al., 2004; Veazey et al., 1998). In the gut, HIV-1 primarily targets resting CD4+ T cells which lack many of the markers associated with activation (Li et al., 2005; Zhang et al., 1999; Zhang et al., 2004). It is thought that this stage of infection is clinically important due to the irreversible destruction of helper T cells and establishment of viral reservoirs in resting T cells (Chun et al., 1998). While acute infection is associated with rapid viral spread and mucosal T cell destruction, the chronic infection is marked by low viral loads and progressive loss of peripheral blood T cells and ultimately AIDS. It is during this phase, which lasts on average 10 years, that CD4+ and CD8+ T cells display high levels of activation (Ascher and Sheppard, 1988). Indeed, many studies have shown increased proliferation of both naïve and memory T cells which leads to abnormally high rate of turnover and/or activation induced death (Fleury et al., 2000; Hazenberg et al., 2000; Hellerstein et al., 2

21 1999; McCune et al., 2000; Mohri et al., 1998; Rosenzweig et al., 1998). Thus, most believe that depletion of T cells during the chronic phase is not caused directly by HIV-1 but rather due to either virus-specific or nonspecific bystander activation. In support of this is data indicating that the degree or level of activation may be the best predictor of disease progression (Leng et al., 2001; Roussanov et al., 2000; Simmonds et al., 1991). Before antiretroviral treatment, HIV infection inevitably led to AIDS and death due to opportunistic infections. Approximately three years after the discovery of HIV-1, the first antiretroviral drug, azidothymidine or AZT, was approved by the FDA for the treatment of HIV-1 infection. This was the first of many nucleoside analog reverse transcription inhibitors (NRTIs) that inhibited the virus and could significantly improve the outcome of AIDS patients. Since 1987 much progress has been made in the fight against HIV-1, although no treatment can completely eradicate the virus. The major antiretroviral drug classes and their mechanism of action are listed in Table 1.1. The three main classes target reverse transcription and viral maturation, however several new targets have been approved for use within the last decade including integration and fusion. Today, ART utilizes a triple combination of drugs as viral mutation can easily overcome the use of one or even two drugs. Despite this effective treatment HIV-1 drug resistance continues to rise and can lead to complete treatment failure. To continue the fight against this virus new drug targets must be identified to expand treatment and combat emerging resistance. 3

22 Table 1.1 Currently available antiretroviral drugs approved for HIV-1 treatment Drug class Mechanism of action Currently available drugs Abacavir, Didanosine, Emtricitabine, Lamivudine, Non-nucleoside reverse Bind directly to the active site Stavudine, Tenofovir transcription inhibitors of reverse transcriptase Disoproxil Fumarate, Zalcitabine, Zidovudine Nucleoside reverse transcription inhibitors Protease inhibitors Fusion inhibitors Integrase inhibitor Mimic normal nucleosides and disrupt new strand synthesis Disrupt viral maturation by mimicking proteases substrates Prevents the conformational change needed for membrane fusion Inhibits strand transfer by binding to integrase Delavirdine, Efavirenz, Nevirapine Saquinavir, Ritonavir, Indinavir, Nelfinavir Mesylate, Lopinavir, Atazanavir Sulfate, Fosamprenavir Enfuvirtide Raltegravir 4

23 1.2 HUMAN IMMUNODEFICIENCY VIRUS TYPE 1 HIV-1 is a member of the lentivirus subfamily of the retroviridae family. All retroviruses are enveloped single stranded RNA viruses. The virus was initially isolated in 1984 as the human T-lymphotropic virus type 3 (HTLV III) and/or lymphoadenopathy associated virus (LAV) (Sarngadharan et al., 1984). Retroviruses are unique when compared to other RNA viruses because these viruses reverse transcribe their RNA into DNA and integrate this DNA into the host chromosome. Thus, retroviruses persist until the infected cell dies. The HIV-1 genome is similar to other retroviruses and contains the typical group specific antigen (gag), polymerase (pol), and envelope (env) genes, and also several regulatory and accessory genes (Figure 1.1). The gag gene codes for a single polypeptide, Pr55 gag that is cleaved by the viral protease (PR) into matrix (MA), capsid (CA), nucleocapsid (NC), and p6. Pol is encoded as a large polypeptide called Pr160 gagpol which is generated by a rare frameshifting event during Pr55 gag translation. This frameshift is caused by a stem-loop structure in the mrna which leads to ribosome slippage in the 5 direction causing a -1 frameshift. This polypeptide is similarly cleaved by the viral PR into the mature proteins integrase (IN), reverse transcriptase (RT), and PR. Env is also codes for a polypeptide (gp160) but is cleaved by host cell proteases during transport to the plasma membrane to yield the surface unit gp120 and transmembrane gp41. In addition to these genes, HIV-1 contains six additional genes; rev, tat, vpr, vpu, nef, and vif. Tat and rev are regulatory genes that are critical for viral transcription and viral RNA export, respectively. The other genes are multifunctional and play major roles in HIV-1 pathogenesis in vivo but may be dispensable for virus replication in vitro. 5

24 Figure 1.1 The HIV-1 genome. Schematic representation of the HIV-1 genome. 6

25 The vpr and vpu genes encode viral protein r (Vpr) and viral protein u (Vpu), respectively. Vpr plays a role in nuclear targeting (Heinzinger et al., 1994) and induces cell cycle arrest (Jowett et al., 1995). Vpu acts in degradation of CD4 in the endoplasmic reticulum through recruitment of a ubiquitin ligase complex and proteosomal degradation (Willey et al., 1992) and proper virus release (Klimkait et al., 1990). The nef gene encodes a protein called negative factor or Nef. Nef plays a role in modulating several cellular pathways including downregulation of both CD4 (Garcia and Miller, 1992; Ross et al., 1999) and major histocompatibility complex I (Williams et al., 2002). Nef also interacts with several signaling proteins in infected cells (Janardhan et al., 2004; Linnemann et al., 2002; Pulkkinen et al., 2004). Lastly, the vif gene encodes the viral infectivity factor protein or Vif. Vif degrades the cellular cytidine deaminase APOBEC3G and prevents hypermutation of the viral DNA (Lecossier et al., 2003; Marin et al., 2003; Sheehy et al., 2003). Together these genes play vital roles for in vivo survival of the HIV THE HIV-1 LIFECYCLE The virus life cycle is typically divided into two distinct phases; the early phase (prior to integration) and the late phase (post integration). Infection starts when the HIV- 1 gp120 binds to the CD4 receptor on the cell surface (Bour et al., 1995). Binding to CD4 induces a conformational change in gp120 which increases affinity for the coreceptors CCR5 or CXCR4 (Dean et al., 1996; Endres et al., 1996). The complex of gp120, CD4, and co-receptor triggers a conformational change in gp41 to induce membrane fusion (Weissenhorn et al., 1997). Membrane fusion allows delivery of the viral core to the host cell cytoplasm. 7

26 The events following entry of HIV-1 are poorly understood. Shortly after entry, the vrna is reverse transcribed in a large nucleoprotein complex (NPC) termed the reverse transcription complex (RTC). It is still unknown if this occurs before, after, or during uncoating of the viral capsid and will be discussed in section Nevertheless, completion of reverse transcription yields a second NPC termed the preintegration complex (PIC). The PIC traverses the cytoplasm and imports the vdna into the nucleus. As with uncoating, the nuclear import process is poorly defined and is currently an area of intense research. Once inside the nucleus, integration of the vdna is catalyzed by the viral IN (Farnet and Haseltine, 1990). The integration reaction occurs in two steps. 1) IN processes the vdna 3 ends by cleaving the two terminal dinucleotides which occurs in the cytoplasm. 2) IN introduces staggered breaks into the host DNA and allows covalent insertion or strand transfer of the vdna into the host cell DNA. The process is complete after cellular repair proteins fill in the gaps between the host and viral DNA to establish the HIV-1 provirus and complete the early phase of HIV-1 replication. The integrated viral DNA serves as the template for HIV-1 gene expression. The 5 -LTR (long terminal repeat) serves as the site for transcriptional initiation and contains numerous features to drive RNA synthesis. The most important of these features is the transactivation response region (TAR) (Berkhout et al., 1989). This stem loop structure is bound by the viral protein Tat to recruit the cellular protein complex known as the positive-transcriptional elongation factor b (P-TEF-b) (Wei et al., 1998). This complex phosphorylates RNA polymerase II and dramatically increases the rate of transcription. 8

27 Transcription of the integrated viral DNA yields unspliced mrnas, partially spliced mrnas, and multiply spliced mrnas (Purcell and Martin, 1993). HIV-1 splicing is a highly orchestrated process that is regulated by both positive and negative cis elements in the viral genome (Tazi et al., 2010). The first mrnas synthesized are multiply spliced and encode Tat, Rev and Nef. During the early phase of HIV-1 mrna expression these fully spliced mrnas accumulate in the cytoplasm but the unspliced and partially spliced mrnas are absent (Emerman et al., 1989; Felber et al., 1989). As Rev accumulates it binds the cis-acting Rev-response element (RRE) located within the env gene region and facilitates the nuclear export of the unspliced/partially spliced mrnas (Maldarelli et al., 1991). Binding of multiple Rev molecules to the RRE allows interaction with the cellular proteins in the exportin 1 (CRM1) RNA export complex (Bogerd et al., 1998) to facilitate export. The final stages of HIV-1 replication include assembly, budding, and maturation. Assembly is mediated by the Gag polyprotein Pr55 gag, which is targeted to the plasma membrane via N-terminal myristolation, multimerizes via Gag-Gag interactions, binds and encapsidates the vrna, and associates with envelope (Ono et al., 2000; Paillart and Gottlinger, 1999). The envelope protein is synthesized and transported through the trans-golgi network, cleaved by host furin proteases (Hallenberger et al., 1992), and glycosylated (Allan et al., 1985). The virus buds from the host cell membrane through recruitment of the endosomal sorting complex required for transport via the domain in the p6 protein of Gag (Dussupt et al., 2009; Sette et al., 2010; Stuchell et al., 2004). During or after the completion of budding the virus undergoes maturation whereby PR 9

28 cleaves the Gag and Gag-Pol polyproteins into the mature viral proteins. This completes the viral lifecycle and the newly formed virions are competent to infect a new target cell. 1.4 LITERATURE REVIEW OF EARLY HIV-1 INFECTION The Viral Nucleoprotein Complexes The mechanisms involved in early HIV-1 infection remain one of the most controversial topics in HIV-1 biology. The process of uncoating, reverse transcription complex/preintegration complex transport and nuclear import are vital to establish HIV-1 infection. Understanding these mechanisms should provide new insights into cellular trafficking processes, the development of improved retroviral vectors for gene delivery, and alternative strategies for blocking viral replication. This study sought to identify cellular proteins involved in these critical processes to determine possible targets for HIV-1 drug therapy. The incoming viral RNA (vrna) is reverse transcribed into double stranded viral DNA (vdna) in the reverse transcription complex (RTC). The newly synthesized vdna is actively transported through the cytoplasm in a large nucleoprotein complex called the preintegration complex (PIC). The RTC is responsible for converting the vrna to vdna while the PIC facilitates nuclear transport and integration of vdna into the host cell genome. These definitions are strictly functional and it is still unknown whether these complexes are biochemically distinct. An initial study of the retroviral complexes of murine leukemia virus (MLV) PIC suggested a complex with an approximate size of 160S that contained all the components necessary for integration (Bowerman et al., 1989). Later several groups identified potential viral components of HIV-1 complexes 10

29 isolated from infected cells. This literature, however, contains conflicting data and there is no firm consensus on the composition of these complexes. The RTC is a filamentous structure of variable size and shape (Chen et al., 1980; Nermut and Fassati, 2003). Data suggests that the viral components of the RTC are reverse transcriptase (RT), integrase (IN), matrix (MA), capsid (CA), nucleocapsid (NC), viral protein r (Vpr) and viral infectivity factor (Vif) (Carr et al., 2008; Chen et al., 1980; Fassati and Goff, 2001). An initial size estimate for the PIC was approximately 320S and identified only the presence of IN within this complex (Farnet and Haseltine, 1991). Later using different methods, other groups demonstrated that in addition to IN, RT, and MA, (Bukrinsky et al., 1993b; Miller et al., 1997) viral nucleoprotein complexes also contained protease (PR) (Karageorgos et al., 1993), Vif (Carr et al., 2008), and Vpr (Heinzinger et al., 1994). The major difference between these reports is the method used to isolate and purify nucleoprotein complexes, indicating the importance for a standard isolation procedure. While these reports identified several viral proteins associated with early NPCs, it remains unclear what critical functions many of these proteins provide during early HIV-1 infection HIV-1 Uncoating The HIV-1 core is a uniquely shaped fullerene cone structure which has a relatively consistent length of nm (Briggs et al., 2003; Hoglund et al., 1992; Welker et al., 2000). The process of uncoating is defined as the dissociation of the CA protein, thus releasing the viral RNA from the viral core. Evidence suggests that proper core stability is vital for reverse transcription, uncoating, and viral infectivity (Forshey et al., 2002). Mutations that increase or decrease core stability reduces viral infectivity 11

30 suggesting that uncoating is a strict process that can adversely affect HIV-1 replication when perturbed. However, the mechanism, timing, and location of uncoating remain difficult to pin down. Several lines of evidence suggest that uncoating occurs rapidly after entry and most of the CA is dissociated from the viral NPC (Bukrinsky et al., 1993b; Fassati and Goff, 2001; Heinzinger et al., 1994; Iordanskiy et al., 2006; Karageorgos et al., 1993). This conclusion is based from the inability to detect significant amounts of CA from isolated NPCs and a perceived lack of viral cores in the cytoplasm of infected cells using TEM. Although, this does not rule out the possibility that experimental techniques may cause premature CA dissociation. A second model proposes that CA core remains throughout reverse transcription and gradually dissociates during cytoplasmic transport leading to a step wise or progressive uncoating (Fassati and Goff, 2001; McDonald et al., 2002; Nermut and Fassati, 2003). Most support for this model comes from the varying size and shape of reported NPCs during HIV-1 infection, although this discrepancy could be a product of the differing methods used to isolate these structures. A recent report suggests uncoating and reverse transcription may be linked (Hulme et al., 2011). In this study, the authors demonstrated that the generation of reverse transcription products coincides with the timing of uncoating. This provides support to the notion that uncoating requires signaling to prevent premature dissociation. A final model suggests that uncoating does not occur until the viral cores reach the nuclear membrane and dissociation proceeds prior to nuclear import. Evidence to support this model indicates integrity and timely disassembly is required for routing to the nuclear compartment (Arhel et al., 2007; Dismuke and Aiken, 2006; Iordanskiy et 12

31 al., 2006; Yamashita et al., 2007). In addition it has recently been shown that the putative nuclear import factor transportin 3 (TNPO3) may interact with CA (Krishnan et al., 2010). Ultimately, the field is divided on the mechanisms and timing of uncoating. Further research of early HIV-1 infection will yield a clearer picture of these poorly understood events HIV-1 Nuclear Import The ability of HIV-1 to successfully infect growth arrested or quiescent cells has spurred extensive research into the process of vdna nuclear import (Bukrinsky et al., 1992; Lewis et al., 1992; Lewis and Emerman, 1994; Naldini et al., 1996; Weinberg et al., 1991). Other members of the retrovirus family such as murine leukemia virus (MLV) are unable to infect non-dividing cells and require breakdown of the nuclear envelope during mitosis to reach the host chromosome (Harel et al., 1981; Miller et al., 1990; Roe et al., 1993). This ability of HIV-1 is important for pathogenesis since infected macrophages are important for dissemination and as a reservoir of virus in infected individuals (Blankson et al., 2002). Even though this phenomenon was observed decades ago, the specific mechanisms of nuclear entry are still a matter of debate. Initial studies examining nuclear import focused on the identification of karyophilic signals in several viral proteins. The MA protein was among the first proteins implicated in nuclear targeting when a transferable nuclear localization sequence (NLS) was identified in its N-terminus (Bukrinsky et al., 1993a). Several years later a second putative NLS was discovered in the C-terminus of MA (Haffar et al., 2000). While these peptides functioned in nuclear targeting when fused to other proteins, MA itself does not 13

32 localize to the nucleus (Depienne et al., 2000). Although these studies proposed an essential role for MA in nuclear import, several studies have observed that MA is dispensable for successful infection (Fouchier et al., 1997; Reil et al., 1998). Thus, it is generally accepted that MA is not the critical protein required for the nuclear import of HIV-1. The accessory protein Vpr has also been implicated in the nuclear import process. Early investigation of this protein demonstrated that it primarily localized to the nuclear matrix during HIV-1 infection (Lu et al., 1993). Subsequent reports indicated Vpr was involved in the nuclear localization of vdna in non-dividing cells, particularly macrophages (Connor et al., 1995; Heinzinger et al., 1994). Additional data indicated that the NLSs in both Vpr and MA contributed to successful nuclear import (Haffar et al., 2000; Heinzinger et al., 1994; Popov et al., 1998b); however several reports did not support this hypothesis (Bouyac-Bertoia et al., 2001; Kootstra and Schuitemaker, 1999; Yamashita and Emerman, 2005) since mutations of Vpr did not drastically alter nuclear import. Vpr interacts with cellular importin and nucleoporins, and accumulates at the nuclear pore complex (Jenkins et al., 1998; Popov et al., 1998a). Additionally, Vpr was found to disrupt the nuclear envelope by inducing herniations that could rupture the nucleus which may mediate an unconventional mode of nuclear import (de Noronha et al., 2001). Although it is unclear whether this nuclear envelope disruption contributes to nuclear import of vdna, this phenomena has been observed in HSV (Scott and O'Hare, 2001) and adenovirus infection (Strunze et al., 2011). HIV-1 IN, which is essential to PIC function, has been implicated in nuclear import and contains several putative NLSs (amino acids , , , 14

33 , and ) (Bouyac-Bertoia et al., 2001; Gallay et al., 1997; Ikeda et al., 2004). However, point mutation at positions 156 and 160 did not alter infectivity or nuclear import, and a mutation at 159 resulted in only a modest reduction in integrated vdna. Overall, this region does not appear to be critical for nuclear import. Mutation of the and NLSs had a severe effect on the association of IN with importin-α and its nuclear import (Gallay et al., 1997). Interestingly, a mutation in the region (R263S) did not alter HIV-1 infectivity or in vitro integration activity, while a mutation at position 262 (R262I) completely rendered the virus noninfectious (Cannon et al., 1996). Thus, it is unclear whether this region is vital for IN nuclear targeting or some other critical protein function. Bouyac-Bertoia et al. identified a highly conserved non-canonical NLS in residues of IN. These residues were reported to be important for the accumulation of IN and/or PICs in the nucleus of infected cells, however subsequent studies failed to solidify the importance of these residues in nuclear import (Dvorin et al., 2002; Limon et al., 2002). Mutations in this area resulted in replication incompetent virus but accumulation of vdna in the nucleus was similar to wild-type suggesting that mutation likely affected IN catalytic activity (Engelman et al., 1995). Thus, the contribution of integrase in the nuclear targeting of vdna has yet to be fully understood. Recent evidence suggests that nuclear import may be regulated by the CA protein (Krishnan et al., 2010; Lee et al., 2010; Yamashita and Emerman, 2004). This idea is supported by evidence which suggests that CA is important for nuclear import via an interaction with the cellular protein transportin 3 (TNPO3). Replacing the HIV-1 CA protein with that of murine leukemia virus produced a virus insensitive to TNPO3 15

34 knockdown, suggesting that CA is the determining factor for nuclear import via TNPO3 (Krishnan et al., 2010; Matreyek and Engelman, 2011). Interestingly, a C-terminally truncated cellular protein polyadenylation factor 6 (CPSF6-358) can bind CA and inhibit HIV-1 nuclear import (Krishnan et al., 2010). However, an interaction between CA and nucleoporins or importins has not been reported, suggesting that the requirement for CA may simply reflect the need for proper viral uncoating prior to nuclear entry (Dismuke and Aiken, 2006). Yet, the role of CA continues to gain traction due to evidence suggesting that it is viral determinant for transportin 3 (TNPO3) dependence. The knockdown of TNPO3 blocks viral replication at the step of nuclear import and will be discussed in more detail below Cellular Proteins Important for Early HIV-1 Infection HIV-1 encodes only 15 proteins, suggesting that throughout early infection there are numerous virus-host protein interactions required for productive infection. Much effort has been devoted to delineate and characterize the structure of HIV-1 NPCs and identify associated host proteins. Previous studies to identify cellular NPC-interacting proteins primarily used yeast two-hybrid or (Kalpana et al., 1994; Violot et al., 2003) immunoprecipitation with integrase (Ao et al., 2007; Cereseto et al., 2005; Cherepanov et al., 2003b; Hamamoto et al., 2006; Jager et al., 2011), or in vitro reconstitution of saltstripped PIC activity using purified or recombinant proteins (Chen and Engelman, 1998; Farnet and Bushman, 1997). While these experiments identified several host factors, they were biased towards integrase alone and/or the process of integration at the expense of other NPC functions such as cytoplasmic transport or nuclear import. A recently published paper has explored a global HIV-human protein interaction using 16

35 affinity tagging (Jager et al., 2011). In this study, the authors tagged and expressed all 18 HIV-1 proteins and polyproteins individually to identify host interacting proteins in two human cell lines (Jurkat and 293T). A total of 497 HIV-human interactions were identified with high confidence. These factors will require further validation and characterization to determine if they are important for HIV-1 replication. Other studies to identify host factors required for efficient infection used whole genome sirna screens (Brass et al., 2008; Konig et al., 2008; Yeung et al., 2009; Zhou et al., 2008). Each of these studies produced over 300 possible proteins, but will again require further analysis to validate if they are important for HIV-1 infection. One caveat to this approach is that the different experimental methods used appear to alter the identified proteins as there was very little overlap (< 5%) between the three studies (Goff, 2008). A summary of the previously identified and validated HIV-1 interacting factors involved in early infection is shown in Table 1.2. Using the yeast-two hybrid system, SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily b, member 1 (SMARCB1, also known as INI 1) was the first integrase binding partner identified (Kalpana et al., 1994). During HIV infection this nuclear protein is transported into the cytoplasm and may have an inhibitory effect on the early steps of HIV-1 replication (Turelli et al., 2001). SMARCB1 association with incoming vdna drastically reduces viral transduction. Another study identified a portion of this protein (S6, amino acids 183 to 294) that is a dominant negative inhibitor of virus assembly that interferes with Gag-Pol/Gag trafficking to the membrane (Cano and Kalpana, 2011; Yung et al., 2001). Recent evidence also indicates that SMARCB1 repress basal HIV-1 promoter 17

36 Table 1.2 Cellular proteins important for early HIV-1 infection Protein Cellular Function Proposed function in HIV-1 infection Barrier to autointegration factor (BAF) Chromatin organization and nuclear envelope assembly Prevents autointegration and associates with PICs gem (nuclear organelle) associated protein 2 (GEMIN2) High mobility group protein A1 (HMGA1) Component of SMN complex and functions in splicing Regulation of inducible gene transcription Interacts with IN. Required for efficient reverse transcription Stimulates integration activity, splice site regulation, and viral transcription Importin α Nuclear transport of proteins Interacts with MA, IN, and Vpr to mediate nuclear import Lens epithelium derived growth factor (LEDGF) Transcription Interacts with IN. Aids in integration by tethering viral DNA to chromatin SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily b, member 1(SMARCB1) Transportin 3 (TNPO3) X-ray repair complementing defective repair in Chinese hamster cells 6 (XRCC6) Transcriptional regulation nuclear import receptor for serine/arginine-rich (SR) proteins DNA repair and V(D)J recombination Associates with PICs. May be required for transcription following integration Binds IN and possibly CA. Required for nuclear import Repairs DNA breaks after integration and protect IN from proteosomal degradation 18

37 activity (Boese et al., 2009). Depletion of SMARCB1 results in an increase in viral replication, but replication was repressed as SMARCB1 expression increased. Barrier to autointegration factor (BAF) prevents suicidal autointegration of HIV-1 PICs in vitro and was shown to co-immunoprecipitate with integrase (Lin and Engelman, 2003). BAF is thought to be involved in several nuclear functions including nuclear assembly and chromatin organization (Gorjanacz et al., 2007; Margalit et al., 2005), as well as gene expression (Dorner et al., 2006; Nili et al., 2001). BAF has been implicated in restoring the integration activity of salt-stripped PICs (Chen and Engelman, 1998), but did not stimulate activity of recombinant integrase in vitro (Carteau et al., 1999). However, BAF may not be important for virus replication. Recent results have shown that sirna knockdown of BAF did not inhibit VSVg-pseudotyped HIV-1 infection (Shun et al., 2007a). Additionally, another study failed to detect the previously reported interaction between MA and BAF (Huang et al., 2011). This study proposes that the BAF-MA interaction is likely due to nonspecific DNA association and may not be biologically relevant. High mobility group protein A1 (HMGA1) is sufficient to restore the activity off salt-stripped PICs but at approximately 500 fold lower than BAF (Farnet and Bushman, 1997). HMGA1 is a involved in regulation of inducible gene transcription and metastatic progression of cancer cells (Cleynen and Van de Ven, 2008). Multiple HMGA1 binding sites have been discovered on the HIV-1 5 LTR suggesting that it may play a role in viral transcription (Henderson et al., 2000). Recent data demonstrated that HMGA1 may be needed to produce the Vpr mrna by alternative splicing (Tsuruno et al., 2011). 19

38 The X-ray repair complementing defective repair in Chinese hamster cells 6 (XRCC6) protein is a component of the non-homologous end joining (NHEJ) pathway, which repairs double strand DNA breaks. XRCC6 is associated with PICs and is responsible for circularization of the vdna to form 2-LTR circles (Li et al., 2001). It may not be strictly needed for integration, but it may function in targeting of the vdna to certain chromatin domains (Masson et al., 2007). Also recent data suggests it is incorporated into virus particles and protects integrase from ubiquitin-mediated degradation upon entry into target cells (Zheng et al., 2011a). Lens epithelium derived growth factor/p75 (LEDGF/p75) is primarily a nuclear protein that is closely associated with chromatin and exhibits a strong affinity for heatshock DNA elements (Singh et al., 2001). LEDGF/p75 has been confirmed to bind integrase by co-immunoprecipitation (Llano et al., 2004) and through a yeast-two hybrid analysis (Emiliani et al., 2005). Moreover, the integration activity in vitro was increased with the addition of recombinant LEDGF/p75 (Cherepanov et al., 2003a). The protein s role in HIV-1 replication is questionable as Llano, et al. did not observe decreased replication in a LEDGF/p75 deficient Jurkat cell line. Interestingly, a recent study using LEDGF/p75 knockout cells indicated that HIV-1 lost its bias for integration into transcription units and conversely had an increased affinity for promoter regions (Shun et al., 2007b). This evidence suggests that LEDGF/p75 may function downstream of the PIC as a targeting factor for integration. The survival of motor neuron (SMN) complex protein gem (nuclear organelle) associated protein 2 (GEMIN2) also interacts with HIV-1 IN. However, unlike the integration factors mentioned above GEMIN2 appears to be required for reverse 20

39 transcription mediated by the interaction with IN (Nishitsuji et al., 2009). This also suggests IN may play a pivotal role in the reverse transcription process. Similar to XRCC6, knockdown of Gemin2 reduced intracellular stability of IN (Nishitsuji et al., 2009). This suggests that IN requires association with cellular proteins for proper stability. Several nuclear pore complex proteins have been identified to interact with PIC components. Vpr, MA, and IN have all been reported to interact with members of the importin-α protein family (Gallay et al., 1997; Gallay et al., 1996; Popov et al., 1998a; Popov et al., 1998b; Vodicka et al., 1998). Importin-α is a protein that forms a heterodimer with importin-β to import NLS-containing cargo into the nucleus. After docking to the nuclear pore complex the importins transfer the cargo into the nucleus in an energy-dependent manner. In addition, it was revealed that Vpr bound to the nucleoporins Pom121 (Fouchier et al., 1998) and NUPL2 (Le Rouzic et al., 2002). Thus, it was reasonable to propose that these interactions were responsible for HIV-1 nuclear import. But none of these appear to be absolutely vital as chimeric viruses replacing IN and MA with their murine leukemia virus counterparts, or completely lacking Vpr, were able to infect non-dividing cells (Yamashita and Emerman, 2005). The next nucleoporin identified was Nup98 based on a finding that it participates in HIV-1 Rev nuclear export (Zolotukhin and Felber, 1999). Experiments using an inhibitor of Nup98 or shrna knockdown was able to significantly reduce HIV-1 vdna nuclear import (Ebina et al., 2004). However, no follow-up studies examining this protein have been reported. A recent study examined several nucleoporins and found that wild-type HIV-1 infection required Nup153 and Nup160, but a mutant virus containing a single mutation in CA 21

40 (N74D) required Nup85 and Nup155 (Lee et al., 2010). The authors suggest that HIV-1 is able to flexibly use alternative nuclear import pathways. Taken together, the mechanism and vital components involved in HIV-1 nuclear import are still a matter of intense debate. Recently, much attention has been focused on transportin 3 (TNPO3) which was one of the genes identified a global RNAi screen for HIV-1 dependency factors (Brass et al., 2008). Initially, a report suggested that this protein was bound to IN and promoted HIV-1 nuclear import (Christ et al., 2008). However, another study suggested that it was CA that was important for TNPO3-mediated nuclear import and not IN (Krishnan et al., 2010). Subsequently, it was suggested that the method of viral entry was a major factor for dependence on TNPO3 since wild-type, but not VSVg envelope still required TNPO3 for successful infection (Thys et al., 2011). This highlights the importance of natural route of entry and may direct future studies to examine wild-type HIV-1 infection. A later report also suggests that CA is the major determinant, but not for nuclear import (De Iaco and Luban, 2011). These authors propose that TNPO3 is required for HIV-1 infection in a step post nuclear entry, as they observed only a decrease in integrated products in TNPO3 depleted cells. Nevertheless, the precise mechanism by which TNPO3 mediates lentiviral nuclear import remains elusive. 1.5 Research Goals There are numerous IN binding/interacting factors, but there is limited information focused on PIC transport and assembly. It is hypothesized that there are still many unknown cellular factors associated and vital for proper PIC/RTC function. The goal of 22

41 my studies was to identify cellular proteins associated with early viral complexes and elucidate the role of these proteins in HIV-1 replication. The first study adapted a unique affinity purification system to study factors associated with viral proteins in early infection. A method was developed to purify functional intact nucleoprotein complexes as an alternative method to identify candidate cellular proteins. Finally, one of these factors was studied extensively and its impact on viral replication evaluated. 23

42 CHAPTER II: IDENTIFICATION OF CANDIDATE HIV-1 INTERACTING CELLULAR PROTEINS BY AFFINITY PURIFICATION OF MATRIX AND INTEGRASE 2.1 BACKGROUND The study of protein-protein interactions has yielded invaluable data regarding host-virus interactions. Peptide labeling and purification of proteins is a useful tool to investigate protein structure and function, and identify protein protein interactions in vivo. A large focus of HIV-1 research is devoted to the identification and characterization of cellular proteins that interact with viral proteins as a means to identify new targets for therapeutic interventions. Many protein-based strategies have been developed for this purpose. Two commonly used purification strategies are single- and tandem affinity purification (TAP). Single-tag protocols are less labor intensive and more efficiently capture target proteins compared to TAP, but exhibit high levels of background. TAP is more specific, but the recovery rate of bait proteins from lysates is quite poor ( 5%) and therefore TAP can miss low abundant proteins. Furthermore, TAP requires two insertions in the same protein which typically increases the chances of protein misfolding and/or inactivation. This study utilized a previously described biotin acceptor sequence (BAS) BirA system for in vivo biotinylation of proteins (Chen et al., 2005; de Boer et al., 2003; Furuyama and Henikoff, 2006; Penalva and Keene, 2004). The BAS is a small 23 amino acid sequence containing a central lysine residue. Co-expression of a BAS tagged protein with the E. coli biotin ligase BirA, catalyzes the specific biotinylation of the central lysine residue of the BAS (Beckett et al., 1999; Schatz, 1993) (Figure 2.1). The 24

43 strength of this system is the specific and strong binding between biotin streptavidin (SA) which is several orders higher than antibody-antigen binding. Additionally, the identification of a minimum 13 amino acid BAS makes this similar to other affinity tag. Purification of biotin tagged proteins can be achieved with a single step under highly stringent conditions. The goal for this study was to identify candidate HIV-1 interacting cellular proteins associated with MA and IN protein complexes. For this study the 19 amino acid BAS was inserted into the C-terminus of HIV-1 proteins MA and IN. MA plays an important role in viral assembly by targeting viral proteins to the plasma membrane (Zhou et al., 1994). MA may also function in early HIV-1 infection as a portion remains associated with the preintegration complex (Bukrinsky et al., 1993b; Lin and Engelman, 2003; Miller et al., 1997). IN is the critical enzyme that catalyzes the insertion of the viral cdna into the host chromosome and may also be important for efficient uncoating (Briones and Chow, 2010). In this study we showed that viral associated MA and IN bearing the BAS were efficiently and specifically biotinylated when BirA was co-expressed in 293T cells. Characterization of these viruses revealed both a replication and infectivity defect in the IN-BAS virus, while the MA-BAS virus exhibited normal replication/infectivity. We observed a loss of integration activity in the IN-BAS clone which was exacerbated by the addition of a biotin moiety; however this clone displayed no defect in reverse transcription. We used both of these clones to affinity purify protein complexes and identified numerous known and unknown cellular proteins associated with IN and MA at 4 hours post infection 25

44 Figure 2.1 Schematic illustrating the BAS-BirA system. BAS-containing proteins are specifically biotinylated on the highlighted lysine (K) residue. The biotinylated protein can then be detected or purified utilizing the biotin-streptavidin non-covalent bond. 26

45 2.2 MATERIALS AND METHODS Plasmids The pnlx plasmid is previously described (Brown et al., 1999) and the pmd2.g VSVg expression plasmid was obtained from the Addgene repository (12259). The BirA expression plasmid pc6bira, and the two BAS molecular clones, NLXIN B and NLXMA B were a kind gift from Dr. Alan Engelman Cell Culture 293TK cells were maintained in Dulbecco s modified eagle medium (DMEM) supplemented with 10% fetal clone 3 (Hyclone, Logan, UT), 8 mm L-glutamine, 100 U/ml penicillin, and 100 ug/ml streptomycin. C and SupT1 cells were cultured in complete Royal Park Memorial Institute medium (RPMI; Invitrogen) supplemented with 10% fetalclone III, 100 U/ml penicillin, 100 µg/ml streptomycin and 4 mm L-glutamine. All cells were cultured in humidified CO 2 incubators (5%) at 37 C. The 293T.BirA cell line was constructed using the following method T cells were seeded into a single well of a six-well dish, grown overnight, then transfected with 2 µg pc6bira using TransIT-LT1 as directed by the manufacturer (Mirus Bio, Madison, WI). Forty eight h post-transfection, the cells were passed into a 10 cm dish and propagated in complete DMEM media containing 20 µg/ml blasticidin S (Invivogen, San Diego, CA). Cells maintained for 2 weeks were then cloned by limiting dilution into 96-well plates, and grown until colonies were visible. Wells containing single colonies were expanded and examined for BirA expression by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting using an anti- 27

46 BirA antibody (Genway Biotech, San Diego, CA), HRP-conjugated anti-chicken IgY secondary antibody (Pierce Biotechnology, Rockford, IL), followed by visualization by chemiluminescence (Pierce Biotechnology). All Western blot images were acquired by flatbed scanning of developed films and, if necessary to improve quality, sharpened and adjusted for brightness/contrast using Adobe Photoshop Virus Preparations Viruses were produced by transient transfection of 293T or 293T.BirA cells. Cells were seeded into 10 cm dishes at 60% confluency. Cells were transfected with 20 µg of viral molecular clone DNA using TransIT-LT1. For pseudotyping with vesicular stomatitis virus glycoprotein G (VSVg), 15 µg of viral molecular clone was premixed with 5 µg of VSVg expression vector pmd2.g (Addgene Plasmid Repository). Media was collected every 24 h for 72 h, clarified by centrifugation at 4000 g for 5 min., and concentrated using Centricon-100 concentrator units as directed by the manufacturer (Millipore, Billerica, MA).Viral stocks were quantified by RT activity using a previously described [ 32 P]TTP incorporation assay (Belshan et al., 2009; Goff et al., 1981) and normalized amounts were used for every experiment. Biotinylated viruses were produced for Western blot analysis by transient transfection as described above. Viral supernatants were harvested at 24 h, clarified, and concentrated by ultracentrifugation through 20% sucrose (w/v in phosphatebuffered saline (PBS)). Afterwards, the media and sucrose were aspirated and virus pellets resuspended directly in 0.25 ml 1 SDS-PAGE sample buffer. For cell lysates, the 293T cells were washed with PBS, harvested, and lysed with 0.5 ml M-PER solution (Pierce Biotechnology). Biotinylated proteins were detected by SDS-PAGE and Western 28

47 blotting using SA-HRP conjugate (GE Healthcare, Piscataway, NJ) and chemiluminescence (Pierce Biotechnology), followed by exposure to film. Viral proteins were similarly detected using antigen specific antibodies followed by species specific HRP-conjugated secondary antibodies (GE Healthcare). For SA-capture experiments, samples were resuspended in 0.5 (virions) or 1ml (cells) RIPB buffer (50mM Tris (ph 7.5)/150mM NaCl/1% NP-40/0.5% sodium deoxycholate/0.1% SDS) and clarified by centrifugation. 50 µl was removed for a pre sample, and then the lysates were incubated with 25 µl SA-sepharose (GE Healthcare) overnight at 4 C. The beads were pelleted by centrifugation and a post sample was removed. The beads were washed in triplicate with 1ml RIPB buffer; bound proteins were released by boiling in 100 µl 1 sample buffer and detected by SDS-PAGE and Western blot. Quantitative imaging was performed using a FluorChem FC2 imaging system (Alpha Innotech Corp., San Leandro, CA) In Vitro Integration Assays PIC-containing lysates were produced using VSVg-pseudotyped HIV essentially as described (Engelman, 2009), except C T-cells were infected by spinoculation (O'Doherty et al., 2000) in the presence of 8 µg/ml polybrene, and the cells were lysed with buffer k (20mM HEPES, ph7.4/150mm KCl/5mM MgCl2/1mM dithiothreitol) containing 0.1% TritionX-100 (Farnet and Haseltine, 1990). Integration activity was detected using nested real-time PCR (Lu et al., 2005) of reaction products with the following modifications. First, linker sequence (5 - ATGCCACGTAAGCGAAACTGC-3 ) was added to the 5 -end of the HIV-specific primer in the first PCR. The linker primer was used in place of one of the HIV primers in the 29

48 subsequent real-time PCR to reduce background HIV amplification, mimicking the nested PCR design developed for analyzing chromosomal integration during infection. Secondly, the vector ptz19r, which is commercially available (Fermentas, Glen Burnie, MD), was used instead of ptz18u/pl as the DNA target during in vitro integration. In vitro integration values were normalized to the levels of viral cdna present in cell extracts. Real-time PCR was performed with an iq5 system and iq SYBR Green Supermix (Bio-rad Laboratories, Hercules, CA) Affinity Purification The BAS/BirA system was used to affinity purify protein complexes for mass spectrometry analysis. First, 1 x 10 8 C cells were infected by spinoculation (O'Doherty et al., 2000) with VSVg pseudotyped HIV-1 containing biotinylated IN (NLXIN B ) or MA (NLXMA B ) and incubated for 4 h at 37 o C. An uninfected cell lysate was prepared in parallel as a negative control. Cells were washed, lysed in RIP-B buffer (50 mm Tris (ph 7.5)/ 150 mm NaCl/ 1% NP-40/0.5% sodium deoxycholate/ 0.1% SDS), clarified, and incubated overnight with streptavidin-agarose (SA) beads. The beads were washed extensively and captured proteins were eluted by boiling then separated on a large format sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS- PAGE) system (Bio-Rad, Hercules, CA). An image of the gel was made by staining the gel with ProtoBlue Safe (National Diagnostics, Atlanta, GA), scanning, and sizing using image software. From the image a grid was constructed and each lane was divided into 20 sections. The gel was overlaid on the grid and each section was excised and stored at -20 o C. Tryptic digestion was performed using the ProteaseMAX surfactant digestion 30

49 protocol (Promega, Madison WI). Following extraction the samples were purified using C-18 ZipTips (Millipore), purified samples were concentrated to ~ 20 μl, and stored at - 20 o C until mass spectrometry analysis Mass Spectrometry Peptide samples were analyzed by the Nebraska Center for Mass Spectrometry using the following methods. Ten micro liters of the extract solution was injected onto a trapping column (300 micron x 1 mm) in line with a 75 micron x 15 cm C18 reversed phase LC column (LC- Packings). Peptides were eluted from the column using a water + 0.1% formic acid (A) / 95% acetonitrile:5% water + 0.1% formic acid (B) gradient with a flow rate of 270 nl/min. The gradient was developed with the following time profile: 0 min 5% B, 5 min 5% B, 35 min 35% B, 40 min 45% B, 42 min 60% B, 45 min 90% B, 48 min 90% B, 50 min 5% B. The eluting peptides were analyzed using a Q-TOF Ultima tandem mass spectrometer (Micromass/Waters) with electrospray ionization. Analyses were performed using data-dependant acquisition (DDA) with the following parameters: 1sec. survey scan ( daltons) followed by up to three 2.4 sec MS/MS acquisitions (60 to 1900 daltons). The instrument was operated at a mass resolution of 8,000. The instrument was calibrated using the fragment ion masses of doubly protonated Glu-fibrinopeptide. The peak lists of MS/MS data were generated using Distiller (Matrix Science, London, UK) using charge state recognition and de-isotoping with the other default parameters for Q-TOF data. Data base searches of the acquired MS/MS spectra were performed using Mascot (Matrix Science, v1.9.0, London, UK). The MSDB database (a comprehensive, 31

50 non-identical protein sequence database maintained by the Proteomics Department at the Hammersmith Campus of Imperial College London which combines entries from TREMPL, SWISSPROT and GENBANK) (Release , 1,942,918 sequence entries) was used and the taxonomy filter was set to human. Search parameters used were: no restrictions on protein molecular weight or pi, enzymatic specificity was set to trypsin, and methionine oxidation was allowed as a variable peptide modification. Mass accuracy settings were 0.15 daltons for peptide mass and 0.12 daltons for fragment ion masses. Significant protein hits that matched more than one peptide with p<0.05 were identified. In the same gel fraction, protein hits matching only redundant peptides with other protein hits of higher scores were removed. Data was compiled from a total of 6 replicates and proteins unique to the infected samples were identified by comparison to the uninfected control. 2.3 RESULTS Characterization of BAS molecular clones To identify cellular proteins associated with HIV-1 in vivo the BAS-BirA system was adapted to label HIV-1 proteins IN and MA. The 19 amino acid sequence was inserted into the C-terminal portion of both IN and MA of the HIV-1 molecular clone NLX (Figure 2.2). To ensure proper retention of proteolytic cleavage of MA, the insertion was placed 5 amino acids upstream of the protease recognition site. To confirm specific biotinylation of IN and MA in vivo we transfected 293T cells with the molecular clones with and without pc6bira. At 24 hours post transfection, the cell culture supernatant was harvested and virus purified by ultracentrifugation through 32

51 Figure 2.2 Construction of BAS-containing IN and MA molecular clones. The canonical BAS, as described in de Boer et al. (2003). The underlined portion defines the common inserted sequence, as indicated by inverted black triangles in. The asterisk denotes the biotin-acceptor lysine residue within the BAS. The BAS in NLXINB is followed by a stop codon after the C-terminal Ser residue in the BAS. NLXMAB harbors the BAS insertion 5 residues N-terminal of the MA-capsid protease cleavage site (indicated by double slash) to permit correct Gag polyprotein processing. 33

52 a 20% sucrose cushion prior to detergent lysis. Biotinylated proteins were separated by SDS-PAGE and detected by Western blot using anti-streptavidin antibodies and IN, MA, and BirA were detected using protein specific antibodies (Figure 2.3). Both biotinylated IN and MA were detected in both the virus and cellular lysates only when BirA was coexpressed. This indicates specific and efficient biotin addition only in the presence of BirA. Detection of IN and MA ensured normal production of virus with or without the addition of biotin and as expected the BAS-containing proteins were slightly larger than the wild-type counterparts. We also detected a small amount of biotinylated full length Pr60 gag in the viral preparations which is indicative of immature virus particles. Next a stable BirA expressing cell line was constructed to efficiently produce large quantities of biotinylated virus and test affinity purification. The 293T cells were transfected with pc6bira and stable cells were produced by selection with Blasticidin S antibiotic. After selection, several stable cell lines with varying degrees of BirA expression were obtained (Figure 2.4A). Cell line D7 displayed normal growth characteristics similar to the parental cells and high expression of BirA and was designated as the primary cell line for subsequent experiments. Next the affinity purification of biotinylated viral proteins was examined in purified virus preps and cell lysates. After transfection with pnlx, pnlxin B, or pnlxma B the cell lysates and viral lysates were affinity purified using SA-agarose beads. Prior to purification a Pre sample was collected and after purification a post sample (Pt) was collected. We observed efficient purification of both IN B and MA B but not wild-type HIV-1 proteins in virion samples (Figure 2.4B). Consistent with their molar ratios, a larger amount of MA was captured compared to IN. Interestingly, we detected a small amount of MA co- 34

53 Figure 2.3 Specific in vivo biotinylation of IN and MA. 293T cells were transfected with NLX molecular clones containing the BAS in IN (pnlxin B ) or MA (pnlxma B ), alone or with the BirA expression plasmid as indicated. Virion lysates (top three panels) were probed by Western blotting using the indicated reagents. The bottom panel indicates the cellular level of BirA expression. 35

54 36

55 Figure 2.4 Affinity purification of IN and MA in stable BirA expressing cells. (A) 293T cells were transfected with pc6bira and selected using blasticidin. Clonal 293T.BirA cell lines were obtained by limiting dilution, expanded, and assayed for BirA expression by Western blotting. (B) SA capture of biotinylated IN and MA proteins from purified virions and cell lysates. Virus was produced by transfection of 293T.BirA cells. Virus and cell lysates were resuspended in RIPB buffer, a pre-sample removed (Pre), and viral proteins captured with SA-agarose beads (AP). After the beads were pelleted, a post-sample was removed (Pt), then the beads were washed extensively with RIPB buffer. Bound proteins were released by boiling in sample buffer, and samples separated by SDS-PAGE were detected using SA-HRP or anti-ma antibody as denoted beneath each blot. Approximate MW sizes are shown to the left of the blots, and viral proteins are indicated on the right. 37

56 purifying with IN (Figure 2.4B, lower right panel), suggesting that these proteins can associate in virus lysates and may also do so in intact virus. We also examined the affinity purification of biotinylated proteins from cellular lysates and detected both IN and MA. We were unable to discern biotin-gag and/or biotin-gag-pol due to a high level of background; however, Pr55 gag and MA were readily detected using an anti-ma antibody (Figure 2.4B lower left panel). As expected, Pr55 gag was purified efficiently in cells expressing biotin-ma, but surprisingly high levels of Pr55 gag were also purified in cells expressing biotin-in. This is likely due to association of gag-pol with gag around assembly sites. Importantly, there was no affinity purification of MA of Pr55 gag in wild-type NLX. Lastly, the observed levels of Pr55 gag were similar in all pre samples, suggesting no defect in viral protein production. pre samples, suggesting no defect in viral protein production. Next, the replication and infectivity of the BAS viruses without biotin was assessed. To examine replication kinetics, virus stocks were produced and normalized by reverse transcription (RT) activity assay. SupT1 cells were inoculated with equivalent amounts of virus and viral supernatant was collected on the days indicated. Viral replication was then measured by RT activity assay (Figure 2.5A). Compared to wild- type NLX, NLXMA B replicated with similar kinetics while NLXIN B displayed a significant delay in replication. This suggested a replication defect due to the insertion within IN. To examine infectivity of both non-biotinylated and biotinylated viruses, single round infections were performed in the MAGI-5 indicator cell line as previously described (Pirounaki et al., 2000). These cells contain a single integrated copy of the β- 38

57 39

58 Figure 2.5 Replication and infectivity profiles of BAS molecular clones. (A) Replication kinetics in SupT1 cells; note the replicating virus is not biotinylated due to the lack of BirA expression in these cells. Results are representative of at least two independent experiments. (B) MAGI-5 cell titer of biotinylated viruses. Virus infectivity was determined after 2 days post-infection by fixing and staining the cells for galactosidase expression. Error bars denote the standard deviation from 9 independent infections. 40

59 galactosidase gene driven by the HIV-1 promoter, thus successful infection leads to constitutive enzyme expression which can be detected by colorimetric assay. These cells were infected with normalized amounts of NLXIN B and NLXMA B with or without biotin. Following fixing and staining with X-gal infected cells were counted to titre viral infectivity (Figure 2.5B). Similar to the above replication data, there was no change in infectivity of NLXMA B with or without biotin. In contrast, the infectivity of NLXIN B was reduced approximately 40% compared to wild-type without biotin and with biotin the infectivity was severely reduced (> 90%). This suggested that the addition of biotin to the C-terminus of IN inactivated the protein and blocked infection. To test the hypothesis that the insertion in IN led to a reduction in specific enzyme activity. We used a previously described in vitro integration assay (Lu et al., 2005). WT and NLXIN B viral PICs were produced in C8166 T cells and integration activity was assayed by real-time PCR (Figure 2.6). This assay revealed a 40% reduction in integration activity in virus without biotin which is in agreement with the data obtained in the MAGI-5 assay. Furthermore, the addition of biotin completely abolished integration activity to the levels of the negative control. These data demonstrated that insertion of the BAS significantly reduced IN catalytic function, while the addition of biotin completely inactivated IN and renders the virus noninfectious. It is well documented that insertions or mutations within IN can lead to pleiotropic effects on HIV-1 replication (Dar et al., 2009; Engelman et al., 1995). Although the loss of integration activity in NLXIN B was not ideal, this study was interested in the complexes prior to integration. Thus to determine if reverse transcription was proceeding normally the production of late reverse transcription (LRT) transcripts of 41

60 Figure 2.6 Biotinylation ablates integration activity. PIC activity of NLXINB produced in the absence and presence of IN biotinylation. Samples were prepared in parallel, and results normalized for levels of cdna synthesis are provided as percentage of activity compared to wildtype HIV-1 NLX. No target ( Target) and uninfected lysate (( ) lysate) controls are shown. Data is representative of two independent experiments. 42

61 both viruses were examined (Figure 2.7). Both BAS containing clones produced higher levels of LRT products compared to wild-type NLX. This indicated that these clones did not have a defect in reverse transcription. The increase in the NLXIN B clone is likely due to accumulation of defective PIC within the cytoplasm. Therefore, we concluded that despite the loss of integration activity in NLXIN B, this virus was adequate for affinity purification of early HIV-1 nucleoprotein complexes Identification of proteins associated with IN and MA proteins complexes in vivo The next step of this study was to isolate IN and MA protein complexes to identify cellular associated proteins. High titer stocks of VSVg-pseudotyped viruses containing IN or MA biotinylated at the C-terminus were produced from the 293T.BirA cell line. C cells were infected by spinoculation and at 4 h post infection the cells were lysed and protein complexes affinity purified with streptavidin-sepharose. For each experiment an uninfected sample was prepared in parallel as a background control. The captured protein complexes were separated by 1-D SDS-PAGE and both IN and MA bands were discernable after staining (Figure 2.8). Additionally, we observed numerous prominent background bands in both the infected and uninfected samples. MS analysis of these bands revealed the isolation of several proteins which utilize biotin as a cofactor for various metabolic reactions. In total, there are five known mammalian proteins that use biotin: Acetyl-CoA carboxylase 1, acetyl-coa carboxylase 2, pyruvate carboxylase, methylcrotonyl-coa carboxylase, and propionyl-coa carboxylase (Pacheco-Alvarez et al., 2002). Our MS analysis revealed that all five were present in our affinity purified samples. 43

62 Figure 2.7 Biotinylation does not affect reverse transcription. NLX molecular clones were produced as described in Fig C8166 cells were infected with equivalent virus amounts and extrachromosomal DNA (HIRT) and cytoplasmic vdna (cyto) was extracted. Late RT vdna products were detected by qpcr and vdna was measured as a ration of cyto/hirt. Error bars denote SEM and data is representative of 3 replicates. 44

63 Figure 2.8 Purification of biotinylated IN and MA protein complexes. Following affinity purification, SA-agarose bound protein complexes were separated by SDS- PAGE. The gel was stained by SYPRO Ruby and imaged at 450 nm. Arrows denote bands representing IN and MA. MW markers are shown on the left side of the gel 45

64 The gels were block excised, each block was trypsin digested, and analyzed by LC MS/MS. IN- and MA-associated proteins were identified by comparing the MS data from the affinity purified samples to the uninfected control samples. As, expected, there were a large number of HIV-1 proteins identified only in the infected samples (Table 2.1). Both IN (Pol) and MA (Gag) were identified by MS and each was present in larger quantities within the corresponding pull-down as indicated by total peptide hits. There were also other HIV-1 proteins in this complex including RT and PR, both of which has been previously identified within early nucleoprotein complexes (Bukrinsky et al., 1993b; Karageorgos et al., 1993; Miller et al., 1997). This further demonstrated specific affinity purification of IN and MA protein complexes. The total list of candidate cellular proteins identified from three independent experiments are shown in Table 2.2. The total protein/peptide hits and total/average Mascot score are given for each protein listed. In total there were 476 proteins identified from these analyses. There were a total of 53 proteins unique to the infected sample and 38 unique proteins in the uninfected control samples. Four putative HIV-1 factors: eef1a1 (EF-tu) (Cimarelli and Luban, 1999), Cyclophilin A (Billich et al., 1995; Colgan et al., 1996; Hammerschmid et al., 1996), Cyclophilin B (Luban et al., 1993), Heat shock protein 90 AA1 (O'Keeffe et al., 2000), and Heat shock protein 70 protein 9 (Agostini et al., 2000; Gurer et al., 2002) were well represented in the infected samples. This screen also identified numerous novel cellular proteins which affinity purified with IN and MA. These proteins represent possible HIV-1 interacting proteins and perhaps perform an essential role in HIV-1 infection. The protein identified with high Mascot score and peptides was the leucine rich pentatricopeptide repeat motif-containing (LRPPRC). This protein became the focus the 46

65 Table 2.1 HIV-1 proteins identified in the infected samples by MS analysis Matrix IP Integrase IP Identification Peptide Hits Total Mascot Score Avg. Score Peptide Hits Mascot Score Avg. Score Gag Pol Gag-pol Matrix Protease Reverse transcriptase

66 Table 2.2 Candidate HIV-1 interacting cellular proteins identified by MS Gene Total rep hits (IN) (MA) Total peptide hits Total Mascot score Mean Mascot score UQCRCP II UQCRCP I MYH Ubiquitin C WASF CaMK I kinase ANT Cofilin Cyclophilin B Mthfd DEHUPA DEHUPB eef-1a1/ef-tu Alpha enolase Beta enolase Gamma-enolase HCD hnrnp F Histone H2A Histone H2B NEFM Obscurin

67 Glutathione peroxidase XTP3TPA HSP70 # LONP HSP90AA MYH HSP70 # Acetyl CoA HADHA SQRDL RPL ATAD3A Tubulin, beta HSP70-9B PCK DUT HMCN LRPPRC Tubulin, beta GSPT MYH TRAP PEX DNPEP CLPX

68 ISOC ACSF RINI RPL CRKL GCDH

69 next research project which will be presented in the following chapter. Other proteins identified were investigated but ultimately the focus was placed on LRPPRC and its role in HIV-1 infection. To validate the presence of the proteins identified by MS we performed Western blots on the affinity purified (AP) protein complexes (Fig. 2.9). The capture of biotinylated IN and MA was confirmed using a streptavidin-hrp antibody (Fig 2.9, top panel). The protein identified with both the highest Mascot score and number of total peptide hits, Leucine rich PPR-motif containing protein (LRPPRC, also known as LRP130), was detected prominently in both the IN and MA AP samples (Fig. 2.9, second panel); however, over-exposure produced a small band in the uninfected sample. Additionally, we validated the presence of two other unique proteins in the IN and MA AP samples, CRKL and eef1a1 (Fig. 2.9, third and fourth panels). Finally, as a control we assayed for a protein ubiquitously detected in our MS analysis, MCCA, that binds endogenous biotin for metabolic processes. As expected, MCCA was detected in all samples (Fig. 2.9, bottom panel). Overall these data confirm several of the MS hits and demonstrate the efficacy of the approach. 2.3 Discussion This was the first study to demonstrate the successful adaptation of the BAS-BirA system to label and affinity purify HIV-1 protein complexes in vivo. The BAS was inserted into MA and IN to yield the HIV-1 molecular clones NLXMA B and NLXIN B, respectfully. Data generated demonstrated site-specific biotinylation of the C-termini of both MA and IN. Biotinylation was only achieved when BAS molecular clones were co- 51

70 Figure 2.9 Validation of candidate proteins in affinity purified samples. Western Blot analysis of affinity purified samples. C8166 cells were infected with VSVg pseudotyped NLXIN B or NLXMA B virus. An uninfected control was performed in parallel. Cells were lysed with IP buffer and protein complexes were affinity purified using SAagarose. SA-agarose bound protein complexes were separated by SDS-PAGE and subjected to Western Blot with streptavidin-hrp and the indicated antibodies. 52

71 expressed with BirA and no biotinylation was observed in wild-type HIV-1 NLX. Both NLXIN B and NLXMA B produced levels of Gag similar to wild-type HIV-1 NLX. Moreover, we detected no defects in Gag processing in either tagged clone. Together these data indicated that both NLXMA B and NLXIN B express and process Gag polyprotein at levels comparable to wild-type HIV-1 NLX. Lastly, a 293T cell line was constructed that stably expresses BirA for efficient production of biotinylated viruses. The MA insertion was well tolerated and the resulting virus replicated similar to wild-type HIV-1. The IN insertion functioned normally in the efferent stages of replication, but examination of afferent steps of infection revealed a significant defect. A more detailed investigation demonstrated a complete loss of infectivity attributable to a loss of integration activity when this virus was biotinylated. Due to this somewhat discouraging result, two different insertions were attempted at positions 212 and 247. These sites were chosen based on previous results which indicated that these positions were receptive to ~19 amino acid insertions without loss of integration activity (Puglia et al., 2006). However, neither of these clones produced infectious virus. Despite the loss of integration, this virus was still utilized to purify complexes from 4 hour infections as this time-point is used to isolate RTCs and PICs prior to integration. This method yielded efficient and rapid purification of biotinylated HIV-1 proteins in vivo. Although the biotin streptavidin bond is very specific, we observed numerous prominent background bands in both the infected and uninfected samples (Figure 2.8). MS analysis of these bands revealed the isolation of several proteins which utilize biotin as a co-factor for various metabolic reactions. The five known mammalian proteins that use biotin are acetyl-coa carboxylase 1, acetyl-coa carboxylase 2, propionyl-coa 53

72 carboxylase, pyruvate carboxylase, and methylcrotonyl-coa carboxylase (Pacheco- Alvarez et al., 2002) and our MS analysis identified all five in our samples. Additionally, there were numerous proteins that appear as non-specific contamination in immunoprecipitation procedures which have been previously documented (Trinkle- Mulcahy et al., 2008). These include several heat shock proteins, myosin, and various cytoskeletal components such as actin and tubulin. Unfortunately, these proteins are inherent to any affinity purification strategy using agarose or sepharose beads and were excluded from further analysis. Taken together, binding of non-specific proteins accounts for a majority of the proteins identified in this analysis. Isolation and tandem mass spectrometry (MS/MS) of these purified protein complexes yielded several known HIV-1 interacting factors as well as numerous unique cellular proteins. In addition, we identified a large number of HIV-1 proteins including MA, IN, RT, and PR in only the infected samples (Table 2.1). This demonstrated efficient capture of target HIV-1 proteins and the specificity of the MS/MS analysis. The known HIV-1 interacting cellular proteins were well represented in our screen and support previously published results. This screen identified several interesting candidate cellular proteins. Several of the identified proteins are associated with cellular functions that may relate to HIV-1 replication. A few proteins will be discussed below with the exception of LRPPRC which will be covered more extensively in the following chapter. The cellular protein TNF receptor-associated protein 1 (TRAP1) is a mitochondrial heat shock protein. This protein plays an important role in protecting cells from undergoing apoptosis (Costantino et al., 2009; Montesano Gesualdi et al., 2007). It was recently identified to interact with sorcin, a protein that also provides a 54

73 cytoprotective function, thus strengthening the role for TRAP1 in apoptotic protection (Landriscina et al., 2010). Of note, another TNF receptor-associated protein called TTRAP was shown to interact with HIV-1 IN and is required for efficient integration (Zhang et al., 2009). Thus it is tempting to speculate that TRAP1 may also be utilized by HIV-1 during infection, possibly to protect infected cells from apoptosis. Another protein of interest is v-crk sarcoma virus CT10 oncogene homolog (avian)-like (CRKL), which we validated by Western blot. This protein kinase contains SH2 and SH3 (src homology) domains which have been shown to activate the RAS and JUN kinase signaling pathways (Ling et al., 1999; ten Hoeve et al., 1993). CRKL also interacts with Rac1 and DOCK2, both of which form a complex with HIV-1 Nef to inhibit chemotaxis and activate T cells (Janardhan et al., 2004; Lu et al., 1996). Thus, it is possible that HIV-1 also utilizes CRKL to modulate signaling pathways during replication. Finally, this study also revealed the candidate interacting protein WAS protein family, member 2 (WASF2). WASF2 forms a multi-protein complex to link kinase receptors with the actin cytoskeleton for cell growth, shape, and motility. One well studied role for WASF2 involves regulation of T cell receptor signaling and T cell activation (Nolz et al., 2006; Nolz et al., 2008). The authors demonstrate that WASF2 is required for T cell receptor-mediated activation of T cells. T cell activation plays an important role in HIV-1 infection and is thought to lead to massive activation-induced death (Fleury et al., 2000; Hazenberg et al., 2000; Hellerstein et al., 1999; McCune et al., 2000; Mohri et al., 1998; Rosenzweig et al., 1998). Thus, one could hypothesize that HIV-1 may utilize the WASF2 complex to induce T cell activation. 55

74 This is the first study utilizing the BAS-BirA biotinylation system to purify HIV-1 protein complexes. This system successfully isolated IN and MA protein complexes from HIV-1 infected cells. We identified several potential HIV-1 interacting proteins which will require further analysis in the future. Despite identifying several possible candidates, this system also produces a high level of background binding which could mask other proteins during MS. Additionally, this study did not produce a fully replication competent HIV-1 IN-BAS molecular clone FUTURE STUDIES These analyses revealed a novel and functional strategy for affinity purification of HIV-1 proteins. Despite this being a powerful technique with high affinity, the resulting background from natural biotin utilizing proteins is disappointing. Several attempts were made to deplete biotin from cultured cells prior to analysis, but unfortunately did not reduce the level of background binding. For this reason, the usefulness of this system for future protein-protein interaction experiments is diminished. 56

75 CHAPTER III: KNOCKDOWN OF THE CELLULAR PROTEIN LRPPRC ATTENUATES HIV-1 INFECTION BACKGROUND Efficient replication of human immunodeficiency virus type 1 (HIV-1) requires interactions with a myriad of host cell proteins. Protein-protein interaction assays, genetic and proteomic screens have identified hundreds of candidate proteins that potentially interact with the virus during productive infection (reviewed in (Bushman et al., 2009)). After entry and uncoating of its viral core, there are many critical steps during HIV-1 replication, including, but not limited to, reverse transcription of the viral RNA (vrna) into DNA, nuclear import of the viral DNA (vdna), and the integration of the vdna into the host cell chromosome, transcription, specific export of unspliced viral mrna, assembly of new virus particles, virion egress, and maturation. All of these steps involve a complex interplay between viral and cellular proteins (Buckman et al., 2003; Lapadat-Tapolsky et al., 1993). HIV-1 matrix (MA) and integrase (IN) are components of the gag and pol genes, respectively. Both proteins are incorporated into virions and present in the HIV-1 reverse transcription and preintegration complexes (Bukrinsky et al., 1993b; Farnet and Haseltine, 1991; Fassati and Goff, 2001; Miller et al., 1997). The functional role of MA in virus assembly is well established. The N-terminal myristolation of MA is critical for targeting the Gag and Gag-Pol polyproteins to the plasma membrane for virus assembly (Ono et al., 2000; Paillart and Gottlinger, 1999). Although MA was among the first viral proteins identified to play a role in HIV-1 preintegration complex nuclear import, its role 57

76 in the early steps of virus replication is controversial. MA is a component of both reverse transcription and preintegration complexes and contains two putative nuclear localization sequences (NLS) (Bukrinsky et al., 1993a); however, deletion of these sequences does not ablate the nuclear import process (Fouchier et al., 1997; Kootstra and Schuitemaker, 1999; Reil et al., 1998). The principle function of IN is facilitating integration of the vdna into the host cell chromosome. IN proteins multimerize at the ends of the newly synthesized vdna and cleave the two proximal nucleotides at each end, resulting in a complex capable of integrating the vdna into a heterologous target. IN contains an NLS, but similar to MA it appears to be dispensable for nuclear import of the preintegration complex (Dvorin et al., 2002; Limon et al., 2002). In the nucleus of cells, IN targets the vdna to sites of active transcription by interacting with the chromosomal tethering protein p75/ledgf (Shun et al., 2007b). In addition to its role in integration, IN also interacts with reverse transcriptase, is required for efficient uncoating (Briones et al., 2010), and reverse transcription (Nishitsuji et al., 2009). Numerous cellular proteins have been identified to interact with MA and IN through in vitro assays. Yeast-two hybrid assays identified HEED (Peytavi et al., 1999), HO3 (Lama and Trono, 1998), and KIF4 (Martinez et al., 2008) as MA-interacting proteins; and integrase interactor 1 (Kalpana et al., 1994), hrad18 (Mulder et al., 2002), YY1 (Inayoshi et al., 2010), and Gemin2 (Hamamoto et al., 2006) as IN-interacting proteins. Three genome-wide RNAi screens have been published to identify HIV dependency factors (Brass et al., 2008; Konig et al., 2008; Zhou et al., 2008). Each screen identified > 200 candidate factors, but there was little overlap between them 58

77 (Goff, 2008). Thus far, four proteins from these screens have been validated as early HIV-1 factors: The transportin 3 nuclear import factor (Krishnan et al., 2010; Rain et al., 2009) which is thought to be important for HIV-1 nuclear import, hnrnpu, a factor identified as a HIV-1 host restriction factor (Valente and Goff, 2009), and two cytoskeletal proteins DNAL1 and MAP4 (Gallo and Hope, 2011). The mass spectrometry analysis of both IN and MA complexes described in Chapter III identified a novel candidate cellular protein, Leucine rich PPR-containing protein (LRPPRC, also known as LRP130). Characterization of LRPPRC found that it interacted with HIV-1 RNA and DNA during early infection. Stable depletion of LRPPRC in target cells reduced HIV-1 and MLV infection. PCR analysis of the early steps of HIV- 1 infection determined that LRPPRC knockdown led to a reduction in nuclear import of vdna and reduced PIC formation. Subcellular fractionation of the stable shlrpprc cells revealed differential knockdown of LRPPRC in various cellular compartments. Analysis of the efferent steps of virus replication found that neither virus release nor viral RNA encapsidation were affected in cells depleted of LRPPRC. The stable knockdown of LRPPRC in 293T cells resulted in delayed cell growth compared to cells expressing a non-specific shrna, suggesting a possible mechanism for the impact of LRPPRC expression on MLV and HIV-1 infection. These data demonstrate that LRPPRC expression is important for efficient HIV-1 replication during the afferent stage. 3.2 MATERIALS AND METHODS Plasmids 59

78 pnlx (Brown et al., 1999), pes (Wan and Ratner, 2001), and psivmne (Henderson et al., 1988) have all been previously described. The shrna plasmids targeting LRPPRC (shlrpprc 02, 03, 04), the non-specific (NS) shrna plasmid, and pflag-lrpprc were obtained from Origene Technologies (Rockville, MD). pnlx-luc, pfb-luc and pcg-gagpol were kindly provided by Alan Engelman (Dar et al., 2009) Cell culture and virus preparation 293T and Hela-CD4-LTR- β-gal cells (NIH AIDS Research and Reference Reagent Program, Germantown, MD) were maintained in Dulbecco s modified eagle medium (DMEM) supplemented with 10% fetal clone 3 (Hyclone, Logan, UT), 8 mm L- glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin. Hela-CD4-LTR-β-gal media was supplemented with 0.1 mg/ml G-418 and 0.1 mg/ml Hygromycin B. Viruses were produced by transient transfection of 293T cells using polyethylenimine (PEI). Briefly, cells were seeded into 10 cm dishes at 60% confluency one day prior to transfection. Fifteen μg of viral molecular clone (pnlx) DNA and 5 μg of vesicular stomatitis virus glycoprotein G (VSVg) expression vector pmd2.g (Addgene Plasmid Repository) were mixed with 45 μl 1mg/ml PEI in Opti-MEM (Invitrogen) minimal media. After a 10 min incubation at room temperature the PEI/DNA solution was added to each dish. HIV- luc viruses were produced by PEI transfection with 10 μg pnlx-luc and 5 μg pmd2.g (VSV envelope). MLV-luc viruses were produced by PEI transfection with 10 μg pfb-luc, 6 μg pcg-gag-pol, and 4 μg pmd2.g. HIV-1 biotin viruses NLXIN B and NLXMA B were produced as previously described (Belshan et al., 2009). To examine incorporation of Flag-LRPPRC into virions, cells were transfected with pflag-lrpprc 48 h prior to 60

79 transfection with pnlx, pes, or psivmne. For all virus stock production, media was collected every 24 h for a total of 72 h, clarified by centrifugation at 4000 xg for 5 min, and concentrated using Centricon-100 concentrator units as directed by the manufacturer (Millipore, Billerica MA). Viral stocks were quantified by RT activity using a previously described [ 32 P]TTP incorporation assay (Belshan et al., 2009; Goff et al., 1981) and normalized amounts were used for every experiment T shlrpprc cells 5x T cells were seeded into a single well of a six-well dish, grown overnight, then transfected with 2 μg shlrpprc 02, 03, 04, or a NS using TransIT-LT1 as directed by the manufacturer (Mirus Bio, Madison WI). Forty eight h posttransfection, the cells were seeded into four 10 cm dishes and propagated in complete DMEM media containing 1 μg/ml Puromycin (Calbiochem, La Jolla, CA) until colonies were visible. Single colonies were isolated using cloning cylinders and seeded into 96- well plates by trypsinization. Clonal cell lines were expanded and LRPPRC expression was examined by SDS-PAGE and Western blot using an anti-lrpprc antibody (H- 300; Santa Cruz Biotechnology, Santa Cruz, CA). The primary antibody was incubated overnight at 4 o C, followed by HRP conjugated anti-rabbit IgG secondary antibody (GE Healthcare, Piscataway, NJ) and visualization by chemiluminescence (Pierce Biotechnology, Rockford, IL). Western blot images were acquired using an Image Station 4000R (Carestream Molecular Imaging, New Haven, CT) and, if necessary to improve quality, sharpened and adjusted for brightness/contrast using Adobe Photoshop. LRPPRC knockdown was quantified by relative optical intensity analysis 61

80 using Kodak image software. The level of LRPPRC knockdown for each cell clone was calculated by comparing the level of LRPPRC expression to the level of actin, which was detected using anti-actin primary antibody (I-19; Santa Cruz Biotechnology) followed by horseradish peroxidase (HRP) conjugated anti-rabbit IgG secondary antibody. Cell lines used for assays were designated 293T -NS, -2.7, -3.6, and Cell fractionations were carried out with the Qproteome kit according to the manufacturer s protocol (Qiagen, Valencia, CA). Western blot of fractionations were carried out as described above. The eif4e protein was detected using antibody FL-217 (Santa Cruz), SLIRP was detected using antibody G-21(Santa Cruz), and c-myc was detected using antibody G-4 (Santa Cruz). Detection of GAPDH was used as a loading control Viral infectivity assays For the luciferase assays, 4x T shlrpprc cells (-2.7, -3.6, -4.1, and -NS) were seeded in triplicate wells of a 24-well plate. The following day each well was inoculated with 100 µl of HIV-luc or MLV-luc and incubated for 48 h at 37 C. Cells were lysed in 100 µl M-PER solution (Pierce Biotechnology) and clarified by centrifugation at 20,000 xg for 5 min. Twenty five µl of the cell lysate was loaded into a white 96-well plate and mixed with 50 µl One-glo luciferase reagent (Promega, Madison, WI). Luciferase activity was measured using a Spectramax L (Molecular Devices, Sunnyvale, CA). Uninfected cell lysate and M-PER solution only were used as negative controls for each assay. Data shown represents at least three independent experiments. For singlecycle infectivity assays, 0.25x10 6 HeLa T4 β-gal cells were seeded into a 6-well plate and transfected the next day with 2 μg of plasmids shlrp -02, -03, -04, or a plasmid 62

81 expressing a non-specific shrna (NS) using TransIT-LT1 (Mirus Bio). 24 h posttransfection, each well of cells was trypsinized and seeded into 4 wells of a 12-well plate. 48 h post transfection, the cells in three wells were infected and assays carried out as previously described (Belshan et al., 2009). The cells in the fourth well were harvested, washed with PBS, lysed with 100 µl of M-PER solution (Pierce Biotechnology, Rockford, IL), and the level of LRPPRC knock-down monitored by Western blot as described above. The results presented represent data from nine independent infections Reverse transcription and nuclear import quantification 293T shlrpprc cells were seeded in 10 cm dishes to achieve 50-60% confluence the following day. NLX-Luc stocks were treated with 2 U/ml Turbo DNase (Ambion, Austin TX) for 1 h at 37 o C. Cells were transduced with normalized amounts of NLX-Luc vector and incubated at 37 o C for 24 h. Media was removed and extrachromosomal DNA was isolated using the modified HIRT protocol (Arad, 1998; Belshan et al., 2009). HIV-1 and cellular DNA was amplified using iq SYBR Green Super Mix on an iq5 multicolor real-time PCR detection system (Bio-Rad) using 250 nm of each primer. Late reverse transcription (LRT) viral DNA was quantified using gag-specific primers NL919 (5 -TTCGCAGTTAATCCTGGACTT-3 ) and NL1054 (5 - GCACACAATAGAGGACTGCTATTGTA). LRT was normalized to detection of mitochondrial DNA using primers MitoFor (5 -ACCCACTCCCTCTTAGCCAATATT-3 ) and MitoRev (5 -GTAGGGCTAGGCCCACCG-3 ). 2-LTR circles were quantified using primers NL500 (5 -AACTAGGGAACCCACTGCTTAAG-3 ) and NL9126 (5-63

82 TCCACAGATCAAGGATATCTTGTC-3 ) and normalized to the level of LRT viral DNA. Heat inactivated virus (30 min at 65 C) was used as a negative control for each experiment ptz19r-ltr The plasmid standard for integration assays, ptz19r-ltr, was constructed by inserting the HIV-1 LTR into ptz19r (Fermentas, Glen Burnie, MD USA). The HIV-1 LTR was amplified by a standard PCR with a 5 BamH1 and 3 Pst1 linkers. The 5 primer was NL1-BH1 (5 -GCGGGATCCTGGAAGGGCTAATTTGGTCC-3 ; BamH1 site underlined) and 3 primer was cnl702pst1 (5 - GCGCTGCAGGCCGAGTCCTGCGTCGAGAG-3 ; Pst1 site underlined). The PCR product was gel purified and digested with BamH1/Pst1. The digested DNA was ligated to BamH1/Pst1 digested ptz19r. The insertion was verified by DNA sequencing (Creighton University Molecular Biology Research Core Facility) Integration assays Integration assays were essentially performed as described (Belshan et al., 2009). Briefly, 2x T cells were infected with concentrated. VSVg-pseudotyped HIV-1 for 6 h. Cells were lysed by hypotonic swelling/dounce homogenization, the nuclei and cell debris removed by centrifugation, and the lysates snap frozen in liquid nitrogen and stored at -20 C. PIC-containing lysates were thawed on ice and treated with 20 µg/ml RNase A (Qiagen) for 10 min at room temperature prior to integration reactions. Reactions were performed in 20 mm HEPES (ph 7.4), 150 mm KCl, 1 mm MgCl 2, 4% 64

83 glycerol, 5 mm DTT. 250 µl of lysate or concentrated fraction was assayed in the presence or absence of 3 ng/µl ptz19r target DNA (Fermentas). The reactions were incubated on ice for 5 min then transferred to 37 C for 45 min. Assays were stopped by the addition of EDTA to 8 mm. Each sample was deproteinated by the addition of SDS and Proteinase K (0.5% and 0.5 mg/ml final concentration, respectively) and incubation at 56 C for 1.5 h. DNA was extracted once with an equal volume of phenol and twice with equal volumes of phenol:chloroform:iaa. The DNA was precipitated with 0.3 M sodium acetate, 1 µl GenElute LPA (Sigma Aldrich, St. Louis, MO USA), and 2.5 volumes 100% ethanol overnight at -20 C. Samples were centrifuged at 21,000 x g for 15 min, washed with 150 µl 70% ethanol, re-centrifuged, air-dried 5-10 min, resuspended in 20 µl Tris-EDTA, and stored at -20 C. The nested PCR strategy to measure integration activity was performed as described previously (Belshan et al., 2009) except that 10-fold dilutions of ptz19r-ltr were used as a real-time PCR standard for each real-time PCR Cell cycle and proliferation assays 293T shlrpprc cell lines were seeded at 1x10 5 cells/10 cm plate in triplicate. At the times indicated the cells were harvested, diluted 1/500 in isotonic saline, and counted using a Z1 Coulter particle counter (Beckman Coulter, Fullerton, CA). Two independent counts were taken for each replicate per time point. For cell cycle analysis, 1x T shlrpprc cells were harvested by centrifugation, resuspended in Vindelov s reagent (10mM Tris (ph 7.6), 10 μg/ml RNase A, 75 μm propidium Iodide, and 0.1% Igepal CA-630) and incubated overnight at 4 o C. Cells were analyzed at the 65

84 Creighton University flow cytometry core using a FACSAria flow cytometer (BD Biosciences, San Jose, CA) and Flowjo software (Treestar Inc., Ashland, OR). JC-1 (5,5,6,6 -tetrachloro-1,1,3,3 -tetraethylbenzimidazolylcarbocyanine iodide) assays were performed using the JC-1 mitochondrial membrane potential assay kit (Cayman Chemical Company, Ann Arbor, MI). MTT ((3-(4,5-Dimethylthiazol-2-yl)-2,5- diphenyltetrazolium bromide) assays were performed on the shlrpprc cells using the CellTiter 96 non-radioactive cell proliferation assay according to the manufacturer s specifications (Promega). A preliminary assay was performed to determine the optimum incubation time. The absorbance values were obtained using a Versamax plate reader (Molecular Devices) Virus release assay and viral RNA quantification shlrpprc 293T cells were seeded in 6-well plates at 50% confluency and transfected with pnlx using PEI as described above. Supernatants were collected at 24 h post-transfection for viral p24 ELISA and viral RNA quantification. p24 ELISAs were performed per the manufacturer s specifications (Advance Bioscience Laboratories Inc, Kensington, MD). viral RNA was isolated from the supernatant using the QIAamp Viral RNA isolation kit (Qiagen) and the relative levels quantified using iscript one-step RT- PCR kit with SYBR green (Bio-Rad, Hercules, CA) using 250 nm of each HIV-1 gagspecific primers NL919 and NL1054. PCR was performed on an iq5 multicolor real-time PCR detection system (Bio-Rad). A dilution series of pnlx DNA plasmid was measured to generate a HIV-1 vdna standard curve for the real-time RT-PCR reaction. Viral RNA encapsidation was calculated as ratio of fg viral RNA/ ng p24. 66

85 Immunoprecipitation Assays For IP-RT-PCR assays of infected cells, 1x 10 6 cells were infected by spinoculation (O'Doherty et al., 2000) with NLX + VSVg virus in the presence of 8ug/mL polybrene. After 4 h the cells were harvested, washed in PBS, and resuspended in 500 µl of IP lysis buffer (Pierce) and pre-cleared with 50 µl protein G agarose beads preequilibrated in lysis buffer for 1 h at 4 C. The beads were removed and the samples incubated with antibody at 1/100 dilution overnight at 4 o C. The following day 50 µl of protein G agarose beads were added and incubated for an additional 4 h at 4 o C. The immune complexes were washed three times for 15 min in lysis buffer and digested for 30 min at 37 C with 100 µg proteinase K in 200 µl of 50 mm Tris (ph 8.0)/1% SDS/10 mm EDTA. Viral RNA was extracted with one round each of phenol-chloroform-isoamyl alcohol (25:24:1), acid phenol-chloroform extraction, and precipitated by isopropanol with 1µl LPA. DNA was extracted once with an equal volume of phenol and twice with equal volumes of phenol:chloroform:iaa. The DNA was precipitated with 0.3 M sodium acetate, 1 µl GenElute LPA (Sigma Aldrich, St. Louis, MO USA), and 2.5 volumes 100% ethanol overnight at -20 C. RNA was purified by centrifugation for 10 min at 13,000 x g, washed with 70% ethanol, and eluted in 50 µl Tris-EDTA. Prior to real-time PCR, the DNA was removed from all RNA samples using turbo DNA-free DNase as directed by the manufacturer (Ambion). Viral RNA and DNA were detected using primers NL919 and cnl

86 3.3 RESULTS Association of LRPPRC with HIV-1 RNA and DNA during early infection but not integrase or matrix LRPPRC was originally discovered as a highly over-expressed transcript in human hepatocellular carcinoma (HepG2) cells (Hou et al., 1994), and has subsequently been described to have several diverse cellular functions. It is a member of a family of proteins that contain numerous copies of a unique structural motif called the pentatricopeptide repeat (PPR) (Small and Peeters, 2000). It was characterized as a RNA/DNA binding protein found in the cytoplasm, mitochondria, and the nucleus of cells (Labialle et al., 2004; Mili and Pinol-Roma, 2003; Tsuchiya et al., 2004). In addition, LRPPRC was identified as a component of the peroxisome proliferator-activated receptor coactivator 1-α (PGC-1 α) complex, which regulates energy metabolism in many tissues (Cooper et al., 2006). LRPPRC is also implicated in the human disease known as Leigh syndrome, a French-Canadian type cytochrome c oxidase deficiency (Mootha et al., 2003; Xu et al., 2004), which suggest an important role in mitochondrial metabolism. Recent work has demonstrated that LRPPRC was a binding partner of eukaryotic initiation factor 4E (eif4e) which aids in translation initiation by recruiting ribosomes to the mrna. However recent work also demonstrated that eif4e regulates the export of mrnas related to cell cycle, including c-myc, Pim-1, and Cyclin D1 (Topisirovic et al., 2009). Additionally, LRPPRC was identified as a strong binding partner of the SRA stem-loop interacting RNA-binding protein (SLIRP) (Sasarman et al., 2010). Interestingly, knockdown of LRPPRC led to a dramatic decrease in SLIRP levels 68

87 and conversely knockdown of SLIRP led to a decrease in LRPPRC levels suggesting interdependent functions (Baughman et al., 2009; Sasarman et al., 2010). To confirm the presence of LRPPRC in RTCs/PICs at 4 hpi and investigate its mechanism of association with HIV-1 NPCs, we performed co-immunoprecipitation (IP) assays with IN and MA, IP-RT-PCR assays for RNA interaction, IP-PCR assays for DNA binding, and IP assays for protein-protein interaction. 293T cells were transfected with Flag-LRPPRC and 24 hours later infected with VSVg-pseudotyped HIV-1. Four hours post-infection the cells were lysed and immunoprecipitated with isotype control, anti-flag, or anti-integrase antibodies. To determine whether LRPPRC interacted with viral RNA or DNA, immunoprecipitates were digested with proteinase K, RNA or DNA extracted, and viral copies quantified by real-time RT-PCR or PCR using gag-specific primers. The mouse isotype control antibody was used to measure background in each experiment and the average background binding was subtracted from each IP sample after qpcr. Infection with heat-inactivated virus used as a negative control and anti-in antibody was used as a positive control for these studies (Allen et al., 1995). Immunoprecipitation of Flag-LRPPRC specifically captured HIV-1 viral RNA (Fig. 3.1A) and DNA (Fig. 3.1B) in early HIV-1 infection. This suggested that LRPPRC associated with viral nucleic acids to associate with RTCs and PICs during early infection. Next, co-immunoprecipitation (Co-IP) assays were performed to determine if LRPPRC directly interacted with IN and/or MA (Fig. 3.2). The 293T cells were transfected with pflag-lrpprc and 24 hours later infected with NLX + VSVg. Cells were lysed and subjected to co-ip with Flag, IN, and MA antibodies. There was no Co- IP of Flag-LRPPRC with IN or MA as detected by Western blot using an anti-flag 69

88 70

89 Figure 3.1. LRPPRC associates with HIV-1 RNA and DNA complexes during early infection (A) Immunoprecipitation RT-PCR in 4h infected cells. 293T cells were transfected with Flag-LRPPRC and infected with NLX. The cells were lysed, immunoprecipitated with the indicated antibodies, viral RNA isolated and quantified by real-time RT-PCR. The background level of viral RNA isolated from an isotype control immunoprecipitation was subtracted from each sample. (B) Immunoprecipitation PCR in 4h infected cells. Experiments were performed as in (A) except viral DNA was isolated and detected by real-time PCR. Data shown in both panels is representative of 3 replicates and error bars show SEM. 71

90 Figure 3.2. LRPPRC does not associate with IN or MA in coimmunoprecipitations. 293T cells were transfected with pflag-lrpprc 24 h prior to infection with NLX + VSVg. Cells were lysed and immunoprecipitations were performed with the indicated antibodies. Western blots were carried out for (A) Flag and (B) integrase, and matrix 72

91 antibody (Fig. 3.2A). Conversely, HIV-1 IN and MA did not significantly associate with Flag-LRPPRC as detected by Western blot using anti-in and anti-ma antibodies (Fig.3.2B). Very small amounts of IN were detected in the Flag-LRPPRC sample, but this was also present in the isotype control suggesting non-specific interaction. Taken together, this data suggested that LRPPRC did not directly interact with IN or MA Cell compartment specific depletion of LRPPRC in 293T cells To investigate the requirement of LRPPRC for HIV-1 infection, we constructed 293T cells stably depleted of LRPPRC by short hairpin RNA (shrna) interference. The cells were transfected with three independent LRPPRC-specific shrna expression constructs and selected by incubation with puromycin. Clonal cell lines were isolated by limiting dilution and screened for LRPPRC knock-down by Western blot analysis. A cell line stably transfected with a nonspecific (NS) shrna expression plasmid was made in parallel for control studies (293T.NS). Several cell clones were identified for each shrna with substantial reductions in LRPPRC expression compared to the NS cells or parental 293T cell line. The cell lines that exhibited the lowest levels of LRPPRC expression for each shrna compared to the NS stable cell line were designated 2.7, 3.6, and 4.1 and used in further experiments. Densitometry analysis of Western blots using actin as a normalization control indicated that compared to the NS control cell line, the 2.7 cells had an approximate 64% reduction in LRPPRC expression and a 99%, and 98% reduction of LRPPRC in cell lines 3.6 and 4.1 cell lines respectively. (Fig. 3.3A). LRPPRC is a multifunctional protein with putative roles in both mitochondrial and nuclear RNA metabolism. To characterize the knockdown of LRPPRC in each cell line 73

92 74

93 Figure 3.3. Stable knockdown of LRPPRC alters the subcellular localization of several proteins. (A) Level of LRPPRC knockdown in 293T cells stably transfected with the indicated LRPPRC-specific shrnas. Knockdown was determined by comparing LRPPRC expression (top panel) to actin expression (bottom panel) and semi-quantified by image ROI analysis. WT shows the parental 293T cell line. (B) Subcellular fractionation of the shlrpprc cell lines. The indicated cell lines were fractionated into cytoplasmic, membrane/organelle, nuclear, and insoluble/cytoskeletal fractions and LRPPRC, eif4e, SLIRP, and c-myc expression were determined by Western blot. GAPDH and Topoisomerase IIα (lower panel) were detected to ensure proper cell fractionation. Data is representative of 3 independent experiments. Data is representative of 2 independent experiments. 75

94 we assessed LRPPRC expression in subcellular fractions. Cells were divided into cytoplasmic, membrane/organelle, nuclear, and insoluble/cytoskeletal fractions and examined by Western blot. No LRPPRC was detected in the cytoplasmic/soluble fraction of cells. The majority of LRPPRC protein was detected in the membrane/organelle fraction consistent with its association with mitochondria. There was only a small decrease in protein expression in this fraction of the 2.7 cell line (Fig.3.3B first panel), while the 3.6 and 4.1 cell lines demonstrated a greater decrease. Both the 3.6 and 4.1 cell lines had a decrease in nuclear-associated LRPPRC, but the 2.7 cell line surprisingly showed no depletion in this compartment. Compared to the NS cell line all three of the shlrpprc cell lines exhibited a strong reduction in LRPPRC expression in the cytoskeletal/insoluble fraction. Combined, these data suggest that the incomplete knockdown of LRPPRC in the 2.7 cell line was due to a lack of substantial knockdown of LRPPRC expression in the mitochondria and nuclei of those cells. LRPPRC interacts with eukaryotic initiation factor 4E (eif4e), an RNA regulatory protein that regulates the expression of several proteins involved in cellular proliferation. It was previously reported that knockdown of LRPPRC reduces nuclear protein levels of eif4e which reduced the expression of several factors associated with cell cycle progression. To determine if LRPPRC knockdown altered the expression of cell cycle factors we examined their expression in the subcellular fractions of the shlrpprc cells (Figure 3.3B). eif4e was expressed in all cell compartments in the NS cells. However there was a decrease in nuclear-associated eif4e in the 3.6 and 4.1 cell lines (Fig. 3.3B second panel), which is in line with previous results indicating that knockdown of LRPPRC inhibits eif4e translocation into the nucleus (Topisirovic et al., 2009). 76

95 Interestingly, the 2.7 cells which have nuclear-associated LRPPRC levels consistent with the NS cells and also exhibited localization of eif4e in the nuclear fraction as expected. Consistent with previous results, SLIRP was detected in only the mitochondrial containing fraction in NS cells, and protein levels were reduced in the 2.7 and 3.6 cell lines, and it was absent in the 4.1 cells (Figure 3.3B third panel). Cellular c- Myc was also detected in NS cells in the membrane/organelle fraction with lower levels detected in the cytoskeletal/insoluble fraction (Figure 3.3B fourth panel). The levels were not altered in the 2.7 cell line, but were reduced in the membrane/organelle fraction of 3.6 cells and absent in the 4.1 cell line. Combined, these data demonstrate that LRPPRC knockdown altered the expression of other cell cycle factors and that the 2.7 cell line has a different knockdown profile compared to 3.6 and 4.1 cell lines LRPPRC knockdown decreases HIV-1 infectivity To test if LRPPRC was necessary for HIV-1 infection we measured the transduction of a HIV-1 virus engineered to express luciferase (HIV-Luc) into the LRPPRC-depleted cell lines. In each experiment the cell lines were seeded in triplicate and inoculated with VSVg-pseudotyped HIV-Luc. Virus transduction was measured by luciferase expression at 48 hours post transduction (Fig. 3.4A). Compared to the 293T.NS cells, an approximately 75% decrease in HIV-Luc transduction was observed for all three of the shlrpprc cell lines, indicating that LRPPRC expression was critical for efficient infection. To determine whether the effect of LRPPRC knockdown was specific to HIV-1, the efficiency of transduction of a murine leukemia virus vector (MLV- Luc) into each cell line was measured (Fig. 3.4B). Interestingly, the 2.7 cell line, which has wild-type levels of nuclear LRPPRC, facilitated MLV transduction at a comparable 77

96 78

97 Figure 3.4. Knockdown of LRPPRC impairs HIV-1 infection. HIV-1 (A) and MLV (B) infection of cells stably depleted of LRPPRC. The indicated shlrpprc cell lines were transduced with normalized levels of VSVg-pseudotyped HIV-1 and MLV luciferase viruses for 48 h. Data is normalized to the NS cells. ** Denotes p< as determined by two-tailed t-test, the error bars denote standard deviation, and the data is representative of at least three independent experiments. (C) Single cycle infectivity assays with LRPPRC-depleted cells. LRPPRC was transiently depleted in Hela-CD4- LTR-β-gal indicator cells with three LRPPRC-specific shrnas (indicated on x-axis) 24 hrs prior to infection with HIV-1. Infectivity was measured by counting blue nuclei (indicated on y-axis). Example of transient knockdown of LRPPRC in Hela-CD4-LTR-βgal cells is shown below the graph by anti-lrpprc and anti-actin Western blots. Data is the average of 9 independent infections and ** denotes p< as determined by two-tailed t-test. Error bars denote SEM. 79

98 level to the NS cells. In contrast, a 50% decrease in MLV transduction was observed in the 3.6 and 4.1 cell lines, both of which have LRPPRC depleted in both the nuclear and cytoskeletal/insoluble fractions. These data indicated that there are possibly two mechanisms through which LRPPRC depletion affected HIV-1 infectivity, and that nuclear LRPPRC is required for retroviral infection. To corroborate the requirement of LRPPRC expression for HIV-1 infection, the effect of knockdown was investigated in a cell line that facilitates CD4-dependent HIV-1 entry. Single-round infectivity assays were performed using HIV-1 indicator HeLa-CD4- LTR-β-gal cells transiently depleted of LRPPRC (Fig. 3.4C). Similar to the stable cells, shrnas 03 and 04 suppressed LRPPRC expression greater than 02. Despite this, all three LRPPRC-specific shrna or the nonspecific shrna were individually transfected into HeLa-CD4-LTR-β-gal cells 24 h prior to inoculation with normalized levels of HIV-1 overnight. Similar to the transduction experiments in the shlrpprc cell lines, all three of the shrnas to LRPPRC significantly decreased HIV-1 infectivity by at least three-fold compared to the cells transfected with the NS control plasmid. These data confirmed that LRPPRC expression was required for efficient HIV-1 infection via its natural route of entry Loss of LRPPRC reduces HIV-1 PIC formation and nuclear import To identify which aspect of infection was impaired in the LRPPRC-depleted cells, we assessed reverse transcription, nuclear import, and PIC formation in the LRPPRC depleted cell lines. Late reverse transcription (LRT) and 2-LTR circle viral DNA products were measured by quantitative real-time PCR to analyze reverse 80

99 transcription and nuclear import, respectively. PIC formation was quantified by in vitro integration assays. To measure viral DNA species each shlrpprc cell line was transduced with equivalent amounts of DNase treated HIV-Luc virus and extrachromosomal DNA was isolated at 24 hpi for qpcr analysis. To control for the extraction of DNA in each experiment the LRT product levels were normalized to the level of mitochondrial DNA in each sample. There was no difference in LRT viral DNA accumulation between any shlrpprc cells compared to the NS cells (Fig. 3.5A), indicating that the depletion of LRPPRC did not affect the steps of virus replication through reverse transcription. Next, the level of viral DNA nuclear import was quantified in each cell line by measuring the accumulation of 2-LTR circles (Fig 3.5B). In these experiments, the levels of 2-LTR circles were normalized to the level of LRT product in each sample to control for infection levels. There was no significant decrease of 2-LTR circle DNA in infected 2.7 cells compared to the NS shrna-containing cells, indicating there was no defect in nuclear import in those cells. In contrast, an approximately 2-fold decrease of 2-LTR circle DNA was recovered from both the 3.6 and 4.1 cell lines, suggesting a defect in nuclear import in those cell lines. To assess preintegration complex (PIC) formation in the LRPPRC depleted cells, each shlrpprc cell line was transduced with equivalent amounts of NLX +VSVg and PICs harvested at 6 hpi. Lysates were normalized for viral DNA content and the level of specific integration activity measured by an in vitro integration assay. All shlrpprc cell lines produced lower levels of PIC activity compared to NS control cells (Fig. 3.5C). The level of PIC activity in the 2.7 and 3.6 cell lines was approximately 2-fold less than the NS cells, and there was a > 4-fold reduction in PIC activity recovered from the 4.1 cell 81

100 82

101 Figure 3.5. LRPPRC depletion reduces HIV-1 nuclear import and PIC formation. (A) LRT product synthesis in LRPPRC depleted cells. HIV-1 LRT products were measured at 24 hpi by real-time PCR and normalized to the level of mitochondrial DNA in each sample. (B) 2-LTR circle accumulation in LRPPRC cells. 2-LTR circles were quantified by real-time PCR and normalized to the level of late RT product in each sample. Data in panels (A) and (B) are from 3 independent experiments and the error bars denote SEM. ** denotes p< 0.01 calculated by two-tailed t-test. (C) PIC activity in shlrpprc infected cells. In vitro integration activity +/- target DNA was determined for each cytosolic lysate at 6 hpi. Data was normalized to the activity recovered from the NS cells and data combined from three independent experiments. The error bars denote SEM and ** denotes p< 0.01 calculated by two-tailed t-test. 83

102 line. These results further demonstrated that LRPPRC depletion affected the early events of HIV-1 infection LRPPRC is not required for virus assembly and release Thus far, the data indicated that LRPPRC was important for the early steps of HIV-1 replication. However, LRPPRC plays a role transcriptional regulation (Labialle et al., 2004) and the nuclear export of mrna (Topisirovic et al., 2009). To investigate whether the expression of LRPPRC affected the efferent stages of virus replication, we measured virus release and RNA encapsidation from shlrpprc cell lines transfected with HIV-1 NLX molecular clone. Supernatants were collected at 48 h post transfection and virus production measured by p24 ELISA. All three of the LRPPRC depleted cell lines produced levels of virus comparable to the NS cells (Fig.3.6A), indicating there was no defect in virus expression, assembly, or release. To evaluate if LRPPRC depletion affected viral RNA encapsidation, the amount of viral RNA in supernatants was quantified by real-time RT-PCR, and the amount per ng p24 calculated (Fig. 3.6B). No substantial change in the levels of RNA encapsidation was observed between the NS and the LRPPRC knockdown cell lines. Combined, this data indicated that LRPPRC was not critical for the efferent steps of virus replication Growth characteristics and expression profiles of the shlrpprc cell lines LRPPRC reportedly modulates cellular proliferation by regulating the expression of factors necessary for the progression of the cell cycle and proper mitochondrial function. Moreover, HIV-1 infection is known to modulate metabolic pathways during productive replication (Chan et al., 2009; Ringrose et al., 2008). Combined these 84

103 85

104 Figure 3.6. Knockdown of LRPPRC does not affect viral RNA encapsidation or virus release. (A) Viral release from shlrpprc cells transiently transfected with pnlx was measured by p24 antigen ELISA. (B) RNA encapsidation in virus produced from shlrpprc cells. RNA was isolated from supernatants and quantified by real-time RT- PCR. Encapsidation was calculated as a ratio of viral RNA (fg) to p24 (ng). Each sample was measured in triplicate and error bars denote SD. Data is representative of two independent experiments. 86

105 findings suggested a possible mechanism for the observed effect of LRPPRC depletion on HIV-1 infectivity. To determine if the LRPPRC-depleted cells had an altered proliferation phenotype, we measured the growth rate of each shlrpprc cell line. Cell lines were seeded at low confluency and growth rate monitored by counting cells at 8 or 16 h intervals for 96 h. Two of the three LRPPRC depleted cell lines, 3.6 and 4.1, displayed a 2-fold reduction in the proliferation rate compared to the 293T.NS control cells (Fig. 3.7A). Indeed, the growth curves correlated with the overall level of LRPPRC knockdown and the expression of nuclear LRPPRC (Fig. 3.3). The partially depleted 2.7 cell line that retained nuclear LRPPRC grew slightly slower than the NS cells. Despite the slower growth, all of the knockdown cell lines still achieved confluency 1-2 days later than the NS cells. Since LRPPRC is implicated in the regulation of mitochondrial metabolism the mitochondrial health of each cell line was evaluated. We first checked general mitochondrial enzymatic activity by MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5- diphenyltetrazolium bromide) assay. Both the 2.7 and 3.6 cell lines produced comparable levels of MTT as the NS cells (Fig. 3.7B). There was a slight reduction in MTT activity in the 4.1 cell line, but it was 2-fold compared to the NS cell line. Second, we measured the mitochondrial membrane potential in each cell line by JC-1 (5,5,6,6 - tetrachloro-1,1,3,3 -tetraethylbenzimidazolylcarbocyanine iodide) dye assay; but there were no differences among the cell lines (Fig. 3.7C). These data suggested that there was no substantial mitochondrial defect in the LRPPRC-depleted cells. Finally, the cell cycle profiles of the cell lines were determined to detect any defects in cell cycle progression for each cell line. Cells were stained with Vindelov s 87

106 88

107 Figure 3.7. Characterization of shlrpprc cells. (A) Growth of shlrpprc cells. NS and shlrpprc cells were seeded in triplicate and counted with a cell counter every 8 and 24 h for a total of 96 h. Data shown is representative of two experiments and error bars denote SD. (B) MTT assay of NS and shlrpprc cells. Cells were seeded in a 96- well plate and were incubated for 24 h at 37 o C prior to addition of the dye reagent. Cells were incubated an additional 2 h and absorbance measured. Assay was performed in triplicate and data is representative of three independent experiments. Error bars denote SD. (C) JC-1 dye assay. Cell lines were incubated in the JC-1 staining solution for 1 h and subjected to flow cytometry. Data shown is representative of three independent experiments and error bars denote standard deviation. (D) Cell cycle profiles of NS and shlrpprc cells. Proliferating cells were stained with Vindelov s reagent and analyzed by flow cytometry to determine the percent of cells in each phase of the cell cycle. Data shown is representative of three independent experiments and the error bars denote SD. 89

108 reagent (Vindelov, 1977) and cell cycle profiles determined by flow cytometry analysis (Fig. 3.7D). Overall there was no significant difference between the three shlrpprc cell lines and the NS cells. The profile of the 2.7 cells was similar to the 293T.NS cells, and the 3.6 and 4.1 cell lines exhibited only a slight increase in cells in the G2 phase complemented by a slight decrease in cells in S phase compared to the 293T.NS cells. The lack of a substantial effect of LRPPRC knockdown on cell cycle profile suggested that cell cycle arrest was not the reason for delayed growth of shlrpprc cells. Together, the analyses of the shlrpprc cell lines indicated that the depletion of nuclear LRPPRC resulted in a decrease in cellular proliferation without a substantial reduction or abnormality in cell cycle phenotype. 3.4 DISCUSSION Retroviral replication is a complex interaction between virus and host factors. These studies identified a new cellular protein, LRPPRC, which is necessary for efficient HIV-1 infection. LRPPRC was identified by mass spectrometry of IN- and MA- protein complexes at 4 hpi, using a biotin-labeling system (Belshan et al., 2009). LRPPRC, a nucleic acid binding protein (Labialle et al., 2004; Mili and Pinol-Roma, 2003; Tsuchiya et al., 2004), was subsequently determined to interact with HIV-1 RNA and DNA during the early steps of replication. Stable knockdown of LRPPRC with three independent shrnas reduced HIV-1 infection during the early phase of virus replication. Infection of all three cell lines resulted in a reduction of PIC formation, and viral DNA nuclear import was impaired in two of the cell lines. LRPPRC expression was not required for efficient virus production or viral RNA encapsidation. Combined these data indicated that 90

109 LRPPRC depletion affects the early events of HIV-1 infection by at least two mechanisms. LRPPRC is a multifunctional protein involved in mitochondrial gene expression and function, cell cycle progression, and RNA regulation. Therefore it was not surprising that knockdown of LRPPRC affected HIV in multiple ways. LRPPRC was detected in the membrane/organelle, nuclear, and insoluble/cytoskeletal fractions, but not the cytosol of cells. Surprisingly, there were distinct patterns of LRPPRC expression and its associated factors in subcellular compartments among the three different shlrpprc cell lines. In all three cell lines LRPPRC was reduced in the insoluble/cyoskeletal fraction, and there was a loss of membrane/organelle associated SLIRP. The 3.6 and 4.1 cell lines displayed a reduction of LRPPRC in the membrane/organelle fraction and both LRPPRC and eif4e in their nuclei. The 2.7 cells exhibited an only a slight reduction of LRPPRC in the membrane/organelle fraction, but showed no change in the expression of eif4e or c-myc. The only defect of HIV replication identified in the 2.7 cells was a loss of PIC formation, which was also observed in the 3.6 and 4.1 cell lines. Thus, cytoskeletalassociated LRPPRC is required for proper PIC formation and represents one mechanism affecting HIV-1 replication. Sequence analysis of LRPPRC identified a putative SEC1 domain involved in cytoskeletal interaction (Liu and McKeehan, 2002). This is supported by the idea that disruption of the cytoskeleton reduces RTC function (Bukrinskaya et al., 1998) and may act similarly for PIC stability through association with microtubules (McDonald et al., 2002). 91

110 Infection of the 3.6 and 4.1 cell lines resulted in an impairment in the accumulation of 2-LTR circles in the nuclei of cells, indicating that nuclear LRPPRC is necessary for efficient nuclear import of viral DNA. Nuclear LRPPRC regulates the expression of mrnas related to cell cycle progression through an interaction with eif4e (Topisirovic et al., 2009). eif4e is a critical regulator of the mrna export of several cell cycle regulatory factors including Cyclin D1, Pim-1, and c-myc (Topisirovic et al., 2009). Knockdown of LRPPRC adversely affects eif4e mediated export due to cytoplasmic retention of eif4e and accumulation of eif4e containing PML-bodies (P-bodies). Consistent with previous studies showing that nuclear LRPPRC regulates eif4e expression (Topisirovic et al., 2009), the level of nuclear eif4e protein was reduced in the 3.6 and 4.1 cell lines, but not the 2.7 cells. Moreover, the 3.6 and 4.1 cell lines grew at a slower rate and exhibited the greatest reduction in the proliferation factor c-myc. The loss of nuclear eif4e and the strong reduction in c-myc levels in these cell lines likely contributed to the loss of proliferation compared to the NS or 2.7 cell lines. Consistent with this, MLV infectivity, which is dependent on cell division, was reduced in the 3.6 and 4.1 cell lines. Additional studies will be needed to determine whether it is the loss of LRPPRC or its associated factors that hinder HIV nuclear import by investigating the loss of eif4e and c-myc independently. The other ascribed functions of LRPPRC provide additional means through which its depletion could impact HIV-1 infection. LRPPRC interacts with both RNA and DNA (Labialle et al., 2004; Mili and Pinol-Roma, 2003; Tsuchiya et al., 2004). HIV replication is regulated extensively at the RNA level and numerous studies implicated RNA processing factors such as the serine/arginine-rich family of proteins as critical for HIV-1 92

111 infection (Jablonski and Caputi, 2009; Lee et al., 2010). These proteins are vital for mrna nuclear export, nonsense-mediated decay, and mrna translation. Moreover, the perturbation of eif4e, which is a critical regulator of other transcription or translation factors including Cyclin D1,and Pim-1, may reduce HIV-1 infection. Although new evidence demonstrates that HIV-1 infection reduces host-cell translation by decreasing eif4e activity without affecting viral translation (Sharma A 2012). This may explain why we observed no significant difference in viral production in our LRPPRC knockdown cells. Finally, the loss of mitochondrial LRPPRC may also have contributed to the reduction of HIV infection. The majority of the data suggests that LRPPRC is primarily a mitochondrial associated protein (Mili and Pinol-Roma, 2003; Sasarman et al., 2010; Xu et al., 2004) and new evidence confirm that it is required for translation of mitochondrial mrnas (Ruzzenente et al., 2011). In addition to proper mitochondrial function and gene expression (Mootha et al., 2003), LRPPRC is essential for COX I and III mrna expression (Xu et al., 2004). Leptomycin B, a specific inhibitor of CRM1 and HIV-1 Revmediated export, also suppresses the export of eif4e mrnas. Thus, in addition to blocking Rev function, it may be likely that the inhibition of CRM1 may inhibit HIV-1 through an eif4e related mechanism. Additional studies will be necessary to investigate the specific role of eif4e in HIV infection. LRPPRC is also critical for proper mitochondrial function and gene expression (Mootha et al., 2003), COX I and III mrna expression (Xu et al., 2004), and phosphoenolpyruvate carboxykinase and glucose-6-phosphate expression (Cooper et al., 2006). LRPPRC-depleted MCH58 fibroblast cells exhibit decreased expression of 93

112 three mitochondrial gene sets: O-glycan biosynthesis, glycosphingolipid biosynthesis, and glycosphingolipid metabolism (Gohil et al., 2010). The effect of perturbing mitochondrial function on HIV-1 infection is not known, but it is clear that virus infection alters mitochondrial function. Two independent proteomic analyses observed the upregulated expression of proteins associated with mitochondria, the TCA cycle, fatty acid oxidation, and oxidative phosphorylation in response to HIV-1 infection (Chan et al., 2007a; Ringrose et al., 2008). Thus LRPPRC reduction could negatively impact HIV-1 infection by altering mitochondrial health and reducing the function of critical metabolic pathways to a level that does not support efficient virus replication. A significant reduction of mitochondrial LRPPRC was observed in the 3.6 and 4.1 shrna cell lines. Despite this, there was no measurable disruption of mitochondrial function in the cells as measured by MTT and JC-1 assays. These results indicate that there was no correlation between the loss of infection and mitochondrial function; however they do not exclude the possibility that mitochondrial associated LRPPRC is critical for HIV-1 infection. Over extended cell passages LRPPRC accumulated in the mitochondrial compartment in both the 3.6 and 4.1 cell lines. This suggests that mitochondrial LRPPRC may have a long half-life in this compartment. Moreover there was a reduction of the LRPPRC-associated protein SLIRP in all three cell lines. SLIRP is important for posttranscriptional mitochondrial mrna expression and consequently oxidative phosphorylation (Baughman et al., 2009; Sasarman et al., 2010). Interestingly, this primarily mitochondrial protein also binds the steroid receptor RNA activator which is an RNA species that acts as a nuclear receptor coactivator (Hatchell et al., 2006). As with eif4e and c-myc, individual knockdown of SLIRP will need to be investigated to 94

113 determine what role it may play in HIV replication. Future examination of SLIRP and additional metabolic indicators and/or a gene array analysis of the shlrpprc cells may reveal new pathways that are critical for efficient HIV 3.5. FUTURE STUDIES A recent examination of the global HIV-1 protein interaction network identified LRPPRC and SLIRP as candidate Vpu-interacting proteins (Jager et al., 2011). Our studies identified this protein with IN and MA complexes and did not test interaction with Vpu. This could explain why we failed to detect a direct interaction between LRPPRC and IN or MA. Vpu is a poorly understood HIV-1 protein. Most studies have focused on Vpu s ability to down-regulate CD4 (Chen et al., 1993; Lenburg and Landau, 1993; Magadan and Bonifacino, 2011; Schubert et al., 1998; Willey et al., 1992) and counteract the viral restriction factor tetherin (Dube et al., 2010; Hauser et al., 2010; Iwabu et al., 2009; Iwabu et al., 2010; Yang et al., 2010). Vpu is characterized as a membrane-associated protein, although it appears to be largely expressed on intracellular membranes such as endosomal compartments and the trans-golgi network. Thus it is hypothesized that Vpu down-regulates these two host cell proteins by disrupting trafficking to the plasma membrane and proteosomal degradation. However, previous studies have not investigated the role of Vpu in early HIV-1 infection. I hypothesize that LRPPRC and SLIRP associate with Vpu during early HIV-1 infection. To test this hypothesis, immunoprecipitations could be performed using the Flag- LRPPRC. The 293T cells would be co-transfected with pnlx and pflag-lrpprc and if reciprocal immunoprecipitations capture Vpu and LRPPRC, this would confirm the Vpu- 95

114 LRPPRC association. Alternatively, fluorescent microscopy could confirm this interaction through co-localization studies. Since both LRPPRC and SLIRP are primarily mitochondrial, co-localization would reveal the cellular compartment in which these proteins associate and may suggest a new functional role for Vpu. To further characterize the interaction, mutational studies could reveal critical interaction domains within each protein. It is possible that the interaction is indirect and mediated by another HIV-1 protein or even nucleic acids. Studies are currently underway to determine the domains critical for LRPPRC association with viral RNA. 96

115 CHAPTER IV: PROTEOMIC ANALYSIS OF EARLY HIV-1 NUCLEOPROTEIN COMPLEXES BACKGROUND The critical early steps of the HIV replication cycle are mediated by two functionally defined nucleoprotein complexes (NPCs), the reverse transcription and preintegration complexes (RTC and PIC, respectively). Upon completion of reverse transcription, the RTC transforms into the PIC, which is operationally defined by the ability to integrate vdna into a heterologous DNA target in vitro (Ellison et al., 1990; Farnet and Haseltine, 1990). The in vitro integration reaction requires only the vdna and IN (Bushman et al., 1990; Craigie et al., 1991); however the large estimated size of the complex (Miller et al., 1997) suggests that these complexes have a complicated composition that includes a variety of viral and cellular factors. The PIC is a delicate complex as studies report inconsistent recovery of viral proteins from PICs, likely due to differences in the method used to purify the complexes. Much effort to delineate and characterize the structure of HIV-1 NPCs has focused on single protein interaction studies. Previous studies to identify cellular NPC-interacting proteins primarily used yeast-two hybrid with integrase (Kalpana et al., 1994; Violot et al., 2003) immunoprecipitation (Ao et al., 2007; Cereseto et al., 2005; Cherepanov et al., 2003b; Hamamoto et al., 2006), and in vitro reconstitution of salt-stripped PIC activity using purified or recombinant proteins (Chen and Engelman, 1998; Farnet and Bushman, 1997; Jager et al., 2011). While these experiments have identified several host factors, they are biased towards integrase and/or the process of integration while 97

116 other NPC functions such as cytoplasmic transport have been largely unexplored. Other studies to identify host factors required for efficient infection have relied on whole genomic sirna screens (Brass et al., 2008; Konig et al., 2008; Rato et al., 2010; Yeung et al., 2009; Zhou et al., 2008). Combined, these studies produced over 300 candidate proteins, but each will require further analysis to validate its importance in HIV-1 infection and determine how it interacts with HIV-1 (Goff, 2008). Biochemical techniques that purify intact and functional complexes will likely be necessary to identify cellular proteins required specifically for the formation and transport of the RTC and PIC. Previous proteomic studies of HIV-1 have focused on virus particles (Chertova et al., 2006; Denard et al., 2009; Saphire et al., 2006), whole cell proteome changes during infection (Chan et al., 2007b; Chan et al., 2009; Melendez et al., 2011; Rasheed et al., 2008; Ringrose et al., 2008) or viral protein expression (Coiras et al., 2006), and biomarker discovery (Ciborowski et al., 2007; Rozek et al., 2007; Wiederin et al., 2009). One report examined the proteome of HIV-1 DNA complexes affinity purified using a biotinylated DNA target (Raghavendra et al., 2010). A recent study analyzed protein interactions with 18 individual HIV-1 proteins using overexpression and affinity purification (Jager et al., 2011). The goal of the studies reported here was to screen intact RTCs and PICs to identify candidate cellular proteins that may contribute to the early steps of HIV-1 infection. To that end, RTCs and PICs were purified by short-length velocity gradient centrifugation, which recovered fully functional complexes as measured by sensitive in vitro assays. The fractions containing functionally active complexes were concentrated and analyzed by label-free RP-HPLC MS/MS. A total of seven biological replicates 98

117 were completed. The total MS data from infected and uninfected samples was compiled and unique and enriched proteins identified. The detection of proteins by Western blot confirmed the accuracy of the MS data. Furthermore, the previously identified cellular protein XRCC6 was confirmed to be associated with HIV-1 DNA. 4.2 MATERIALS AND METHODS Cell culture and viral infection 293T cells were maintained in Dulbecco s modified eagle medium (DMEM) supplemented with 10% fetal clone 3 (Hyclone, Logan, UT), 8 mm L-glutamine, 100 U/ml penicillin, and 100 ug/ml streptomycin. C T cells were cultured in RPMI 1640 media supplemented with 10% fetal clone 3 (Hyclone), 8 mm L-glutamine, 100 U/ml penicillin, and 100 ug/ml streptomycin. PICs were produced using methods previously described (Belshan et al., 2009; Engelman, 2009). HIV-1 molecular clone NLX (Brown et al., 1999) was produced by transient transfection of 293T cell using Polyethylenimine (PEI). Viral supernatant was harvested, filter concentrated, and stored at -80 o C. Virus was treated with 100 U/ml Turbo DNase (Ambion, Austin, TX USA) for 60 min at 37 C prior to infecting cells. For each individual infection 1x10 8 C cells were inoculated with 1 ml concentrated virus at an MOI 25 in a 2 ml final volume of complete media containing 16 µg/ml polybrene. The mixture was transferred to a six well plate and incubated for 10 min at room temperature to allow the cells to settle to the bottom of each well to produce an even layer of cells during spinoculation. Spinoculation was performed by centrifugation at 1050 x g for 2 h and 20 C using a biosafety rotor (O'Doherty et al., 2000), and cells were immediately placed in a tissue 99

118 culture incubator for 1 h. Afterwards 2 ml of complete RPMI was added to each well, the cells transferred to 75 cm 2 tissue culture flasks, and brought up to a volume of 42 ml. At 20 hpi, the cells were pelleted at 250 x g for 4 min at 4 C. The media was discarded and the cells washed twice with 5 ml Buffer K(-) (20 mm HEPES (ph 7.5), 150 mm KCl, 5 mm MgCl 2 ). For hypotonic lysis, the cells were resuspended in 1.8 ml cold hypotonic buffer (20 mm HEPES ph 7.3, 5 mm KCl, 1.5 mm MgCl 2, 1 mm DTT, and 1X Protease Inhibitor Cocktail IV (EMB Biosciences, Gibbstown, NJ)), and incubated on ice for 20 min with occasional mixing by inversion. The cells were transferred to a pre-cooled dounce homogenizer and lysed by 40 strokes with a tight pestle. The nuclei were pelleted by soft centrifugation (1000 x g for 5 min), the supernatant transferred to a fresh tube, and the lysates further clarified by centrifugation at 21,000 x g for 5 min. The supernatants of replicate infections were combined and separated into 550 µl aliquots that were flash frozen in liquid nitrogen and stored at -80 C until purification Velocity gradient centrifugation and fractionation Gradients were made in 14 x 89 mm open-top polyclear centrifuge tubes (Seton Scientific, Las Gatos, CA) which were pre-rinsed thrice with millipure water and dried. Gradients were formed utilizing the Gradient Master Station as directed by the manufacturer (BioComp, Fredericton, NB, Canada). After formation the gradients were kept on ice. A 500 µl volume was removed from the top of each gradient and 500 µl of lysate applied drop-wise. Ultracentrifugation was performed with an LE-80 Ultracentrifuge (Beckman Coulter, Fullerton, CA USA) using slow acceleration and slow deceleration. The fractions were collected using the piston fractionator on the Gradient 100

119 Master Station using cold Buffer K (-) as a wash buffer. The fractions were collected at the speed of 0.5 mm/sec, distance 3.95 mm, for a total distance of mm. The collection line was rinsed with Buffer K (-) between each fraction. Fractions were collected on ice during fractionation and stored at 4 C. The linearity of gradients was monitored by measuring the refractive index of a 15 µl sample of each fraction using a refractometer. Gradient fractions were concentrated using 3,000 MWCO column concentrating filters at 4 C as directed by the manufacturer (Millipore). A combined 10.0 ml of each fraction plus 1.0 ml of PBS was added at a time to each column. After centrifugation the flow through discarded, the column refilled with lysate/pbs and recentrifuged. The fractions were concentrated to a volume less than or equal to 500 µl. The concentrated sample was transferred to a 1.6 ml tube and the filter rinsed with 100 to 200 µl PBS to recover residual protein. Each sample was brought up to a final volume of ~650 µl with PBS. The refractive index of each sample was measured and if the concentration was > 5% (w/v) sucrose the samples were returned to the column, diluted with PBS, and re-centrifuged. This process was repeated until each sample contained < 5% (w/v) sucrose Viral DNA quantification and integration assays DNA was isolated from samples (lysates or gradient fractions) using PureLink PCR Purification Kit (Invitrogen) with a vacuum manifold (Qiagen, Valencia, CA USA) according to manufacturers protocol except that the DNA was eluted using 50 µl of Tris- EDTA pre-warmed to 65 C. DNA samples were stored at -20 C. Ten-fold dilutions (100, fg/µl) of pnlx plasmid in 3 ng/µl trna were used as a PCR standard. Therefore the presented values are relative to the amplification of the plasmid standard. 101

120 Real-time PCR reactions contained 1X IQ SybrGreen Supermix (Bio-Rad, Hercules, CA USA), 250 nm of primers, and 2 µl of DNA sample or standard. The primers used to detect late reverse transcription (LRT) product vdna were NL919 (5 - TTCGCAGTTAATCCTGGCCTT-3 ) and cnl1054 (5 - GCACACAATAGAGGACTGCTATTGTA-3 ) corresponding to nucleotides of NL4-3 (Genbank # M19921); and the primers used to detect early reverse transcription (ERT) product vdna were NL497 (5 -GCTAACTAGGGAACCCACTGCTT-3 ) and cnl- 574 (5 -ACAACAGACGGGCACACACTAC-3 ) corresponding to nucleotides of NL4-3. Real-time PCR was performed using an iq5 Multicolor real-time PCR detection system (Bio-Rad). The PCR parameters were: an initial denaturation for 6 min at 95 C, then 38 cycles of denaturation at 95 C for 20 sec, annealing at 55 C for 20 sec, and extension at 72 C for 30 sec. The real time data capture occurred during the extension step. DNA extraction efficiency was monitored by quantifying mitochondrial DNA levels using primers and amplification parameters described previously (Butler et al., 2001). DNA extraction efficiency was monitored by quantifying mitochondrial DNA levels using primers and amplification parameters described previously (Butler et al., 2001) In-solution tryptic digest Protein levels from concentrated fractions were measured using NanoDrop ND (Thermo Fisher Scientific, Inc., Waltham, MA). Approximately 10 µg of protein were added to 15 µl of digestion buffer (50 mm ammonium bicarbonate), 1.5 µl of reducing buffer (100 mm DTT), and 0.5 µl water. The samples were heated to 95 o C for 5 minutes and allowed to cool. Next 3 µl of akylation buffer (100 mm Iodoacetamide) was added and the sample was incubated for 20 minutes. Finally, 1 µl trypsin (1 µg/µl) 102

121 (Promega, Madison, WI) was added and incubated at 37 o C for 3 hours. An additional 1 µl trypsin was added and the sample was incubated overnight at 30 o C. After tryptic digest samples were purified using C18 Zip Tips (Millipore, Billerica, MA) according to manufacturer's procedure Protein identification by nano-lc-ms/ms Samples were re-suspended in 0.1% formic acid in water prior to LC-MS/MS analysis. Protein identification was performed as described previously (Rozek et al., 2007) using ESI-LC-MS/MS system (LTQOrbitrap, Thermo Scientific, Inc., San Jose, CA) in a nanospray configuration using a microcapillary RP-C18 column (New Objectives, Woburn, MA) for fractionation. The spectra were searched using Sequest search engine in Bio- Works 3.2 software (Thermo Scientific Inc., San Jose, CA) using the following parameters: threshold for Dta generation = 10000, precursor ion mass tolerance = 1.4, peptide tolerance = 2.00 and fragment ions tolerance = Database NCBI.fasta from was used with two missed cleavage sites allowed and at least two peptides were required for protein identification Data parsing and ingenuity pathway analysis Data obtained from Sequest was parsed using ProteoIQ software (NuSep, Bogart, GA) and Proteome Discoverer (Thermo Fisher Scientific). Data parsing Proteome Discoverer was performed using the Fasta database (ftp://ftp.ncbi.nih.gov/blast/db/fasta) and the following parameters: Precursor mass tolerance = 10ppm, Fragment mass tolerance =: 0.8 Da, average precursor mass = false, fragment mass = true, dynamic modification = oxidation (M), and static 103

122 modification = Carboxymethyl (C). Data parsing using ProteoIQ was performed under default parameters using the UniRef90 database (UniProt Consortium). The peptide filter was set with a maximum of 1 peptide per protein and a minimum probability of The protein filter was set to 4 spectral counts per identification as previously described (Old et al., 2005). Pathway analysis was performed using Ingenuity Pathway Tool software (Ingenuity Systems, Inc, Redwood City, CA). The ProteoIQ parsed data sets were loaded into the software and highest scoring pathways were obtained using the default parameters Western Blots Concentrated protein fractions were TCA precipitated, washed twice with 100% acetone, and resuspended in sample buffer. Samples were separated by SDS-PAGE and transferred to PVDF using a semi-dry transfer cell (Bio-Rad). XRCC6/Ku70, RPS6, hnrnp L, ILF2, ILF3, PRKRA, and Annexin A6 were detected using anti-ku70 (E-5) antibody, anti-rps6 antibody (C-8), anti-hnrnp L antibody (4D11), anti-ilf2 antibody (A-8) (Santa Cruz Biotechnology, Santa Cruz, CA), anti-ilf3 antibody (AJ1402a; Abgent, San Diego, CA), anti-prkra antibody ( ; Proteintech Group Inc, Chicago, IL), and anti-annexin A6 antibody (31026; Abcam, Cambridge, MA) followed by HRP conjugated anti-rabbit or anti-mouse IgG secondary antibody (GE Healthcare, Piscataway, NJ) and visualization by chemiluminescence (Pierce Biotechnology, Rockford, IL) Immunoprecipitation-PCR 104

123 C cells were infected with equivalent amounts of NLX + VSVg by spinoculation. The virus pellet was resuspended in 300 µl of IP lysis buffer (Pierce) and pre-cleared with 50 µl protein G agarose beads (GE Healthcare) pre-equilibrated in lysis buffer for 1 h at 4 C. The beads were removed and the samples incubated with antibody at 1/100 dilution overnight at 4 o C. The following day 50 µl of protein G agarose beads were added and incubated for an additional 4 h at 4 o C. The immune complexes were washed three times for 15 min in lysis buffer and digested for 30 min at 37 C with 100 µg proteinase K in 200 µl of 50 mm Tris (ph 8.0)/1% SDS/10 mm EDTA. Viral DNA was extracted with one round each of saturated phenol, phenol-chloroform:iaa extraction, and precipitated by 100% ethanol with 1µl linear polyacrylimide (LPA). DNA was purified by centrifugation for 10 min at 13,000 x g, washed with 70% ethanol, and eluted in 50 µl Tris-EDTA. Viral DNA was detected using primers NL919 and cnl RESULTS AND DISCUSSION Optimal infection parameters To identify cellular co-factors of HIV-1 infection it was critical to isolate the maximal quantity of viral complexes while avoiding any disruption. To determine the timepoint at which the greatest number of PICs were present in cells, we infected C cells by spinoculation with cell-free, VSVg-pseudotyped, replication competent HIV-1 NLX concentrated by ultrafiltration and treated with DNase. Infected cell lysates were harvested at 4, 6, and 18 hours post infection to determine when the highest level of viral DNA and PIC activity could be recovered (Figure 4.1, A and B). Late reverse transcription (LRT) product DNA was quantified by real-time PCR using 105

124 106

125 Figure 4.1. Assessment of infection length and extraction method. (A) Recovery of LRT vdna at various times post infection. DNA was isolated from cell lysates obtained by hypotonic swelling and dounce homogenization at the indicated times post infection. DNA extractions were normalized by quantification of mitochondrial DNA (data not shown) and viral DNA levels quantified by real-time PCR using gag-specific primers. H.I.= control infection with virus heat inactivated at 65 C for 30 min. Data is representative of two independent experiments. (B) Total level of PIC activity recovered by hypotonic swelling and dounce homogenization at various times post-infection. The levels (fg) of integrated products were determined by real-time PCR for reactions performed in the presence (+) and absence (-) of target DNA. PICs were harvested from infected C cells at the times indicated; (-)= control reaction with uninfected lysates; Nev.= reaction with lysate from cells treated with 20 mm nevirapine. Data shown is representative of three independent experiments. (C) Recovery of vdna from lysates extracted using the indicated methods. A single infection was split into equivalent amounts, DNA isolated, and LRT vdna quantified by real-time PCR. Plot is representative of three independent experiments. (D) Total integration activity recovered from lysates extracted by the indicated methods. Data is representative of three independent experiments. In all experiments the error bars denote the standard deviation of the representative experiment. 107

126 primers specific to the MA region of the gag gene. The highest levels of vdna were recovered at 18 hours post infection (hpi) as compared to either 4 or 6 hpi. Control infections with heat-inactivated virus (H.I.) yielded undetectable levels of vdna, demonstrating the absence of plasmid DNA contamination. PIC activity in lysates was measured using a sensitive in vitro integration assay (Belshan et al., 2009; Lu et al., 2005). The assay measures integration into a plasmid target DNA by nested PCR. In vitro integration activity was measured for each sample in the presence and absence of target DNA. Specific integration activity was detected at all time points. However, consistent with the viral DNA levels, greater than ten-fold higher levels of integration activity were observed in lysates harvested at 18 hpi compared to those harvested at 4 or 6 hpi. Control reactions with uninfected lysates and lysates generated from cells treated with the reverse transcriptase inhibitor nevirapine demonstrated the specific extraction of PIC activity. Combined, these data demonstrated that the maximal amounts of PICs were recovered at 18 hpi. As noted above, the recovery of viral proteins from the NPCs varied by extraction methods. Moreover, HIV-1 viral cores are sensitive to miniscule amounts of detergent (Kewalramani and Emerman, 1996; Liska et al., 1994; Wyma et al., 2000; Yu et al., 1993). To avoid dissociating the complexes, we investigated harvesting the NPCs by mechanical disruption without detergents versus detergent lysis (0.1% TritonX-100 or 0.01% NP-40). To do this we performed parallel 18 h infections and measured the recovery of viral DNA and PICs. Infected C cells were split into three samples, which were lysed by either hypotonic swelling/dounce homogenization, incubation with 0.1% TritonX-100, or incubation with 0.01% NP-40. Despite comparable levels of vdna 108

127 in the extrachromasomal samples, at least four-fold higher levels of LRT vdna were consistently recovered from cells harvested by hypotonic swelling/dounce homogenization (Figure 4.1C). In parallel experiments the lysate samples were measured for total PIC activity (Figure 4.1D). Analogous to the LRT vdna levels, twoto four-fold higher levels of PIC activity were recovered from the cells lysed by hypotonic swelling/dounce homogenization as compared to the cells lysed with mild detergent. Notably, normalization of the levels of PIC activity to the amount of vdna in each sample indicated no statistical difference in PIC activity/fg vdna between the different lysis methods, indicating the increase in the hypotonic samples resulted from recovery of a higher quantity of PICs instead of differences in activity among the extraction methods. Combined, these data show a maximal recovery of PICs at 18 hpi. The use of replication competent virus suggests that the high level production of RTCs and PICs likely resulted from cell-to-cell infection following a robust primary infection. The method to partially purify HIV-1 RTCs and PICs using velocity gradient centrifugation is summarized in Figure 4.2. For each biological replicate infected and control uninfected samples were prepared in parallel. To obtain enough protein for each biological replicate, a total of ~ 1.6x10 9 cells were infected in sixteen 100 ml cultures using the spinoculation technique (O'Doherty et al., 2000). The lysates were clarified by centrifugation, snap frozen in liquid nitrogen, and stored at -80 C until gradient purification. HIV-1 NPCs were purified using 5-45% sucrose gradients (w/v in buffer K) centrifuged for 1 h at 41,000 rpm. Gradients were fractionated from the top to the bottom by positive displacement. The sedimentation pattern of each replicate was monitored by quantifying the levels of LRT viral DNA in each fraction. The typical 109

128 Figure 4.2. Schematic depiction of the methods used in this analysis. C8166 T- cells were infected with HIV-1 + VSVg pseudotyped virus for 20 hours. Cells were lysed by hypotonic swelling and dounce homogenization. Lysates were loaded on a 5-45% sucrose gradient and subjected to ultracentrifugation for 1 hour at 207,570 x g. Gradients were fractionated and digested in solution with trypsin prior to MS/MS analysis. For each infected replicate, a parallel uninfected sample was subjected to the same procedure. 110

129 sedimentation pattern of vdna, shown in Figure 4.3A, was a single peak of DNA from approximately fraction To determine if the vdna containing fractions contained PICs we assayed the fractions for in vitro integration activity. To do this, replicate fractions from six gradients were combined, concentrated, and buffer exchanged to < 5% sucrose. Each combined fraction was assayed for integration activity and the level of specific integration activity calculated as a ratio of the activity in the presence and absence of target plasmid. Therefore a value > 1 indicated specific integration activity and a value 1 indicated no specific activity. The mean specific activity for the fractions is shown in Figure 4.3B. The highest level of PIC activity was recovered from fractions 15-17, and lower levels were recovered from fractions PIC activity was not recovered from fractions These results demonstrate the presence of functional PICs in fractions of 1 h 5-45% sucrose gradients. To determine the location of RTCs in gradients, we measured the level of endogenous RT activity in the fractions. The assays were performed using a previously described method that monitors the synthesis of the first product of reverse transcription, the strong stop or early RT (ERT) vdna product (Nermut and Fassati, 2003). The assay was performed as described previously, but adapted to utilize realtime PCR to detect ERT synthesis, and parallel reactions were assembled and incubated on ice were run as a control and confirm the recovery of specific activity (Fig. 4.3C, gray plot). High levels of ERT vdna synthesis were detected in fractions 12-16, demonstrating the purification of functional RTCs. Combined, the data shows that PICs and RTCs co-sediment in 1 h 5-45% sucrose gradients. In addition, these data suggest 111

130 112

131 Figure 4.3. Isolation of functional HIV-1 preintegration complexes and reverse transcription complexes. (A) Sedimentation of HIV-1 vdna complexes in 5-45% (w/v) sucrose gradients run for 1 h. Fractions were obtained by piston displacement from the top (fraction 1) to the bottom of the tube. The fraction number is indicated at bottom of each graph and fraction 20 represents the irretrievable portion in the round bottom of the tube. DNA was isolated from each fraction and the vdna quantified by real-time PCR. Plot represents greater than 3 independent infections and fractionations. The error bars denote the standard deviation of the single real-time PCR. (B) Integration activity recovered from the fractions of 1 h 5-45% sucrose gradients. For each experiment six replicate lysates were centrifuged, fractionated, combined, and concentrated. The concentrated fractions were individually assayed for PIC activity. Integration activity was determined by calculating the integrated product in the presence and absence of target DNA. The bars in the graph present the combined mean PIC activity for the fractions indicated. A ratio > 1 indicates IN activity (dashed line). Error bars denote the combined standard error of the mean for the data of least three independent experiments. (C) Endogenous RT (EndoRT) activity was measured by in vitro strong stop DNA synthesis as quantified by real-time PCR using ERT vdna primers. Result from a parallel mock reaction held on ice is show by gray plot. 113

132 that this purification method does not dissociate proteins critical for proper function of these complexes. Thus, hypotonic lysis recovers the largest amount of complexes without disrupting their activity. The NPC isolations and purifications for proteomic analysis were performed a total of seven times (seven independent cultures and infections), which were considered biological replicates. The level of infection and yield of complexes was measured for each experiment and found to be consistent, likely due to the use of the transformed T- cell line versus donor-derived peripheral blood mononuclear cell cultures. Prior to proteomic analysis each replicate was analyzed for vdna sedimentation and all replicates produced a single broad peak of vdna spanning fractions similar to Figure 4.3A. For each biological replicate, sucrose gradient fractions 11 to 19 were trypsin digested and analyzed separately by nano-rp-lc-ms/ms. MS spectra were separately submitted for database searches using Proteome discoverer and ProteoIQ Proteome Discoverer For each biological replicate infected and control uninfected samples were prepared in parallel. A total of ~ 1.6x10 9 C cells were infected using the spinoculation technique (O'Doherty et al., 2000). The lysates were clarified by centrifugation, snap frozen in liquid nitrogen, and stored at -80 C until gradient purification. After velocity gradient centrifugation the sedimentation pattern of each replicate was monitored by quantifying the levels of LRT viral DNA in each fraction. All replicates produced a single broad peak of vdna spanning fractions similar to Figure 4.3A. The infections and NPC purifications for proteomic analysis were performed a total of seven times (seven 114

133 independent cultures and infections), which were considered biological replicates. The level of infection and yield of complexes was consistent across replicates, likely due to the use of the transformed T-cell line versus donor-derived peripheral blood mononuclear cell cultures. For each biological replicate, equivalent protein levels of sucrose gradient fractions 11 to 19 were trypsin digested and analyzed separately by nano-rp-lc-ms/ms. MS spectra were separately submitted for database searches using Proteome Discoverer. Results of database searches were exported to the excel file. Next, data were combined; first fractions into one biological replicate and then seven biological replicates to control and infected groups. These two combined files were compared to each other based on the Sequest score. In total, the analysis by Proteome Discoverer produced 49,417 peptide hits, 25,079 in control fractions and 24,338 in infected fractions (Figure 4.4A). A total of 11,055 proteins were identified from the peptide pools of both the infected and uninfected samples. Comparison of total number of peptides or proteins identified in the infected or uninfected samples indicated a near 1:1 ratio over the seven biological replicates, with a minute bias toward the control fractions (Table 4.1). These data suggest the absence of any bias in the comparison of control and infected MS data due to differences in the input levels of protein between the infected and control groups. To identify the proteins composing RTCs/PICs we assumed that proteins unique to the HIV-1 infected cells would likely represent viral interacting proteins. Therefore, the protein hits common to both the control and infected data sets were excluded to identify candidate RTC/PIC proteins. Despite the large pool of proteins identified, there were less than 90 unique proteins in the infected samples (Fig 4.4B). These were ranked by 115

134 116

135 Figure 4.4 Venn diagrams of proteins identified in the Proteome Discoverer analysis. The total Sequest TM data obtained using Proteome Discoverer was organized and sorted using Microsoft Excel (A) Total number of peptides identified in each sample after data sorting. (B) Venn diagram indicating the number of unique proteins in both the infected and control samples. 117

136 Table 4.1. Proteome Discoverer Protein Hit Summary Protein Hits % Infected 5, % Control 5, % Total 11,

137 total Sequest probability score and are presented in Table 4.2. The protein with the highest total peptide hits was XRCC6 (P12956, XRCC6_HUMAN) which belongs to Ku70 family (UniProt KB). This protein functions in the non-homologous end joining pathway and was described previously as a HIV-1 interacting protein.(li et al., 2001) XRCC6 is incorporated into virus particles and protects integrase from degradation upon entry into target cells (Zheng et al., 2011b). Other HIV-1 interacting proteins identified in the infected samples include: HSP70 (Agostini et al., 2000; Gurer et al., 2002; Iordanskiy et al., 2004), TFRC (Madrid et al., 2005), NONO (Raghavendra et al., 2010), and TAF6 (Ou et al., 1994; Zhou and Sharp, 1995) (Table 4.2), all of which interact with HIV-1 during infection. Thus the data was in agreement with previously published results. Proteins that were not reported previously as components of viral complexes included: FCRL6, GRIN2C, RAP18, MARCKS, and GPR107. Also proteins involved in RNA metabolism: STK33 and NONO were identified as unique candidate HIV-1 interacting proteins. NONO, identified in a previous proteomic screen (Raghavendra et al., 2010), is a RNA-binding protein that functions in transcriptional regulation (Dong et al., 2009), RNA splicing (Kameoka et al., 2004), and double strand DNA repair (Bladen et al., 2005), and is considered to be a primarily nuclear protein. NONO interacts with genomic RNA from the Hepatitis delta virus (Sikora et al., 2009) and the heterodimer of NONO/PSF can bind HIV-1 mrna instability elements to regulate viral expression (Zolotukhin et al., 2003). The search results also uncovered numerous proteins that appeared to be enriched in the infected fractions (Table 4.3). Arbitrarily, we noted any protein that had 6-fold or greater peptide hits in the infected sample compared to the uninfected sample as 119

138 Table 4.2. Selected proteins unique to infected samples 1 NCBI GI # Gene name 2 Independent Protein Identifications Total Peptide Hits XRCC6 (Ku70) gi FCRL6 Fc receptor-like glutamate receptor, ionotropic, N-methyl D-aspartate 2C NONO non-pou domain containing, octamer-binding; p54(nrb) MARCKS protein MAX-like protein X Golgi matrix protein 130 (GM130) FAM167B family with sequence similarity 167, member B Serine/threonine kinase 33 (STK33) Total Score PDIA3 protein disulfide isomerase family A, member 3 1 Certain cytoskeletal and ribosomal proteins were omitted due to the high concentration in both infected and control samples. 2 NCBI Gene name. Unknown protein hits were identified by BLAST homology search. 120

139 Table 4.3. Selected proteins enriched in infected samples NCBI GI # , , , Gene name Independent Protein Identifications (Infected/Control) Annexin A6 12 / 6 55 / HIST1H2AH histone cluster 1, H2ah 16 / 1 83 / CKLF-like MARVEL transmembrane domain containing SFRS protein kinase 1 8 / 1 22 / 4 Total Peptide Hits (Infected/Control) Total Score (Infected/Control) / / / 1 16 / / / MAST4 3 / 1 8 / / Lung cancer oncogene / 1 16 / / AR4/FMR2 family, member 1 4 / 1 12 / / MAP1A 4 / 1 14 / / RAPGEF2 Rap guanine nucleotide exchange factor (GEF) 2 3 / 1 13 / / Brain abundant, membrane attached / 3 / 1 93 / 25 signal protein Protein identification based on BLAST search as protein record noted as unknown 121

140 enriched. These may represent proteins which are normally present in particular fractions, but actively migrate to the HIV-1 nucleoprotein complexes upon infection. One protein that was highly enriched in the infected sample was annexin A6 (AnxA6). AnxA6 is a member of the annexin protein family which are dynamic and multifunctional calcium and membrane-binding proteins (Moss and Morgan, 2004). AnxA6 is the largest member of the annexin family and research suggests it functions in cholesterol homeostasis, membrane trafficking/structure, actin-cytoskeleton, and signal transduction (see (Enrich et al., 2011) and references therein). Another annexin family member, A2, was implicated in HIV-1 infection as a factor required for viral production in monocyte-derived macrophages (MDM) (Ryzhova et al., 2006). A subsequent study confirmed the importance of A2 in particle production in MDMs but this role was not recapitulated in other cell types (Rai et al., 2010). Therefore A2 may be a macrophagespecific host cell factor that regulates HIV-1 infectivity. Published data suggest that AnxA6 expression is closely linked to caveolae formation in various cell types (Cubells et al., 2007). Of note, the major caveolae protein caveolin-1 can inhibit HIV-1 infection through transcriptional repression mediated by NF-κB (Wang et al., 2011). Thus it is tempting to speculate that HIV-1 may alter caveolae homeostasis to achieve productive infection. Lastly, two additional datasets were generated by this analysis. One dataset included proteins that were unique to the control fractions. There were a total of 121 unique proteins identified in these fractions and could represent cellular factors that are mislocalized or down-regulated by HIV-1 infection. A summary of the proteins with the highest score is presented in Table 4.4. Finally, there were also a number of proteins 122

141 Table 4.4. Selected proteins unique to control samples NCBI GI # Gene name Poly(A) binding protein, cytoplasmic 1 (PABPC1) Eukaryotic translation elongation factor 2 Independent Protein Identifications Total Peptide Hits Total Score BASP hnrnp K UPF SRSF hnrnp M Poly(A) binding protein, cytoplasmic 4 (PABPC4) Calmodulin hnrnp L Certain cytoskeletal and ribosomal proteins were omitted due to the high concentration in both infected and control samples. 2 NCBI Gene name. Unknown protein hits were identified by BLAST homology search. 123

142 also enriched in control fractions (Table 4.5). As noted above, we used an arbitrary cutoff of 6-fold greater number of peptide hits to identify these proteins. Of note, two related proteins were identified in this dataset, ILF2 and ILF3. These two proteins form a heterodimer that functions in gene expression and stabilize mrnas (Kiesler et al., 2010; Pfeifer et al., 2008). Interestingly, the ILF3 protein interacts with HIV-1 Rev and could inhibit HIV-1 mrna export by altering the intracellular localization of Rev (Urcuqui- Inchima et al., 2006; Urcuqui-Inchima et al., 2011) Validation of the extraction method Because the mechanical extraction conditions posed some risk of producing nonspecific interactions we confirmed the interaction of a known candidate component XRCC6 (Ku70), which was unique in the infected samples. To confirm the presence of XRCC6 in the proteomic samples, Western blots were performed on the concentrated fraction samples (Figure 4.5A, top panel). Ribosomal protein S6 (RPS6), which was identified at similar levels in both the infected and control fractions by MS, was used as a control in these assays (Figure 4.5A lower panel). Consistent with the MS data, RPS6 was detected abundantly in all fractions in both the infected and control samples. In agreement with the Proteome Discoverer data, XRCC6 was observed in fractions of the infected samples, but nearly undetectable in the parallel control fractions. Moreover, this was consistent with previous results that demonstrated XRCC6 was associated with PICs (Li et al., 2001; Zheng et al., 2011b). To confirm that XRCC6 was directly associated with HIV-1 NPCs immunoprecipitation-pcr (IP-PCR) assays were performed. Infected C cells were lysed and immunoprecipitated with isotype, anti-xrcc6, or anti-integrase antibody. The immunoprecipitates were digested 124

143 Table 4.5. Selected proteins enriched in control samples NCBI GI # Gene name PRKRA protein kinase, interferoninducible double stranded RNA dependent activator Independent Protein Identifications (Infected/Control) Total Peptide Hits (Infected/Control) Total Score (Infected/Control) 1 / 5 2 / / STAU1 protein 1 / 3 2 / / YBX1 Y box binding protein 1 1 / 5 6 / / , interleukin enhancer binding factor , 90kDa 1 / 6 1 / / interleukin enhancer binding factor 2, 45kDa 1 / 10 4 / / PRPF40A PRP40 pre-mrna processing factor 40 homolog A 1 / 4 1 / / Heterogeneous nuclear ribonucleoprotein F 1 / 3 2 / / Heterogeneous nuclear ribonucleoprotein C 3 / 11 7 / / ClpX caseinolytic peptidase X homolog 1 / 4 3 / / family with sequence similarity 120A 1 / 6 2 / / Protein identification based on BLAST search as protein record noted as unknown 125

144 126

145 Figure 4.5. XRCC6 is present in the infected samples and is associated with HIV-1 DNA. (A) Proteomic fractions were TCA precipitated, resuspended in sample buffer, and separated by SDS-PAGE. Western blot was performed using the antibodies to XRCC6 and RPS6. The Cell Lys. sample denotes a C8166 whole cell lysate. (B) C8166 cells were infected with NLX + VSVg and incubated 20h prior to lysis. Lysates were co-immunoprecipitated with the antibodies indicated above. DNA was isolated and viral DNA was detected using gag-specific primers. HI denotes heat-inactivated virus. ** denotes p < 0.01 as calculated by two-tailed T test. Data represents three independent replicates 127

146 with proteinase K and DNA isolated by phenol:chloroform extraction. The vdna was detected by real-time PCR using HIV-1 gag-specific primers. The anti-integrase IP was the positive control and heat-inactivated virus and an isotype control antibody group were used as negative controls in these experiments. XRCC6 immunoprecipitated HIV- 1 DNA in infected cells at levels comparable to HIV-1 integrase (Figure 4.5B), confirming its association with PICs early after HIV-1 infection. These data demonstrate the capability of the experimental design to isolate and identify a factor known to interact with HIV-1 NPCs Validation of Proteome Discoverer candidate proteins As with XRCC6, Western blot analysis was performed on the fraction samples to validate the presence of candidate proteins identified by Proteome Discoverer (Figure 4.6). To validate the proteomic results, several proteins were detected in fractions from both infected and control fractions. Once again, the RPS6 protein was used as a control and was readily detected in all fractions (left column top panel). Annexin A6 (Anx6) was enriched in the infected fractions according to the MS results. As expected, we detected higher levels of Anx6 in the infected fractions compared to the control fractions in replicate experiments (left column second panel). This confirmed that Anx6 is enriched in HIV-1 NPC-containing fractions. Next, we examined two heterogeneous nuclear ribonucleoproteins (hnrnps), K and L, Both hnrnps were identified uniquely in the control fractions. Both hnrnp K and L proteins were detected prominently in the control samples, although small amounts were observed in the infected fractions (left column lower panels). This demonstrated that these hnrnps were enriched in the control samples and confirms the MS results. 128

147 Figure 4.6. Validation of candidate proteins identified by Proteome Discoverer. Hypotonic gradient fractions were TCA precipitated, resuspended in sample buffer, and separated by SDS-PAGE. Several candidate proteins were detected using proteinspecific antibodies. A whole C8166 cell lysate was used as a positive control to ensure proper detection of proteins 129

148 Lastly, we probed for several additional proteins that were enriched in the control fraction data set: protein kinase, interferon-inducible double stranded RNA dependent activator (PRKRA), interleukin enhancer binding factor 3, 90kDa (ILF3), and interleukin enhancer binding factor 3, 45kDa (ILF2) (Fig. 4.6). All three proteins were enriched in the control fractions which further validated the proteomic results. Taken together, these western blots demonstrate the efficacy of the MS analysis and provide several candidate proteins for further characterization ProteoIQ Surprisingly, the data obtained from the ProteoIQ analysis differed significantly from the data obtained by Proteome Discoverer. The total analysis identified 8,340 proteins in both the infected and uninfected samples. Comparison of total number of proteins identified in the infected or uninfected samples indicated a near 1:1 ratio suggesting that over the seven biological replicates all the samples contained similar levels of protein (Fig. 4.7A). The ProteoIQ results identified 99 proteins unique to the infected samples and 86 in the uninfected samples with 8155 proteins shared between both (Fig. 4.7B). Overall these data demonstrated the absence of any bias in the MS data due to differences in the input levels of protein between the infected and control groups. The different parameters and database likely contributed to the discordant values obtained by ProteoIQ when compared to Proteome Discoverer. The unique candidate proteins based on normalized spectral count from ProteoIQ are shown in Table 4.6. Several known HIV-1 interacting proteins were identified including GANAB (Gruters et al., 1987; Ratner et al., 1991), CD58 (Swingler et al., 2003), VPS37B (Stuchell et al., 2004), BTRC 130

149 131

150 Figure 4.7 Venn diagrams of proteins identified in the ProteoIQ analysis. The complete Sequest TM data set was parsed using ProteoIQ software. (A) Total number of proteins identified in each sample after data parsing using default parameters. (B) Venn diagram indicating the number of unique proteins and shared proteins in each sample. 132

151 Table 4.6 Unique proteins identified in infected fractions Sequence Id Gene Total Score Total Peptides Total % Seq Coverage Total Spectra Normalized Spectral Counts Q15323 KRT Q9P2W3 GNG B4DPZ0 PTTG1IP Q9HAT2 SIAE B5MCN3 SEC14L B0QZD1 GPSM Q9BS14 GANAB B3KSC5 SLBP E9PSI6 AKIRIN Q92520 FAM3C Q96S27 Gene X D6RCM1 RBPJ Q6ZQU3 cdna FLJ46889 fis, clone UTERU B4DKF3 cdna FLJ51882, Q6NT90 ATP13A Q8TER1 WITHDRAWN P29508 SERPINB Q8NBG7 HCG Q96LV5 INTS4L Q6ZUU3 C3orf B4DWY7 ADAM Q15937 ZNF Q9H715 cdna FLJ21558 fis, clone COL P84157 MXRA Q14228 EHS

152 B4DYL6 TESK B4DU58 CAPG P19256 CD F5H465 TMEM120B Q9NU98 RPS B4DNA9 cdna FLJ P57060 RWDD2B E7EWC6 POLR2J E9PI23 TMEM Q9H1R2 DUSP Q9Y657 SPIN B7Z3S6 cdna FLJ B4DF36 cdna FLJ E5RIB8 KIAA Q96DN5 WDR B4DTN8 TMEM39B Q6ZSZ7 cdna FLJ45097 fis, clone BRAWH C9JHH5 TEX Q5JTA8 MAGT B4DIP4 EIF2AK O15198 SMAD E7ETU6 SMAP C9JWM2 SCAMP A6QRH8 FAM3A Q2F839 HSPA Q969F8 KISS1R B4DF39 cdna FLJ Q9H521 LOC D6RIS6 PGM

153 Q3BBV2 NBPF E9PI43 Uncharacterized protein Q9NT46 DKFZp434H Q59F78 RGS B1ANH7 Putative uncharacterized protein C1orf B0QYL9 POLR2F Q06520 SULT2A C9J2Y4 FAM136A F5H1F6 VPS37B E7EMN6 PPP1R Q6ZW34 LOC Q9P0F4 HSPC A4D2H4 HINT Q6ZNX1 CDNA FLJ26957 fis, clone SLV Q6NSB4 HP D6RJB4 DGKQ Q6MZG8 DKFZp686I B3KVX7 cdna FLJ41700 fis, clone HCHON Q53RS4 ABI O43459 SALL Q7Z614 SNX B1AMA8 MAPKAP D3DPU0 HCG A8KAJ9 TYW C9J539 GLB B7Z3Q5 cdna FLJ Q9UHK3 Chromosome 5 open reading frame Q8TDS3 PGR B4E055 ABCB

154 A8K857 cdna FLJ C9JZJ2 BRCC Q8NH40 OR6S Q9UF24 DKFZp586K Q86WI5 BTRC E7ETG6 PPFIA E7ESM3 EIF3M Q8NF47 FLJ00349 protein B4DEG4 HNRNPM C9JPP7 C1orf87 chromosome 1 open reading frame P32320 CDA Q59GE3 MHC II, DR alpha variant Q6L9N2 UMODL Q96RT5 TSC A8MQU8 CD F6S878 C2orf68 chromosome 2 open reading frame

155 (Besnard-Guerin et al., 2004; Evrard-Todeschi et al., 2006; Margottin et al., 1998), and CD28 (Ott et al., 1997; Swigut et al., 2001; Venkatachari et al., 2007). The remaining proteins currently have no biological connection to HIV-1 biology and thus represent new HIV-1 candidate interacting proteins (CIPs). Next, ingenuity pathway analysis (IPA) was utilized to determine if the identified candidates participate in a specific cellular network. This analysis revealed 5 pathways which contained at minimum 10 CIPs (Fig. 4.8). The highest scoring interaction network identified by IPA is involved in cell death, inflammatory response and cell signaling (Fig. 4.8A). We identified 18 CIPs (highlighted in red) that function in this cellular pathway and several of these act on the NFκB, AKT, ERK1, and IL1. All of these proteins are modulated by HIV-1 infection through viral proteins (Kino et al., 2005; Olivetta et al., 2003; Qiao et al., 2006). ERK1 phosphorylates the HIV-1 PIC associated proteins MA and Vif which is thought to be an important modification for HIV-1 infectivity (Bukrinskaya et al., 1996; Yang and Gabuzda, 1998; Yang and Gabuzda, 1999). The second highest scoring network pathway identified functions in cell to cell interactions (Fig. 4.8B). In total there were 12 CIPs present in this cellular pathway. Several proteins in this pathway are influenced by the tumor protein 53 (TP53) which has several connections to HIV-1 infection. TP53 binds directly to Nef and this interaction is thought to protect T cells from TP53-mediated apoptosis (Greenway et al., 2002). Additionally, TP53 has been shown to interact with RT and increase the fidelity of reverse transcription (Bakhanashvili, 2001; Bakhanashvili et al., 2004). Interestingly, a member of the disintegrin and metalloprotease domain (ADAM) family of proteins, ADAM2 is involved in this pathway. Recent evidence identified ADAM10 as a necessary 137

156 138

157 Figure 4.8 The top two scoring pathways identified by ingenuity pathway analysis. (A) The cell death, inflammatory response, and cell signaling pathway. (B) The cell-to-cell interaction pathway. Proteins unique to the infected samples are indicated in red and relative increased or decreased abundance is indicated by increasing intensity of red or green, respectfully. Circles containing an inner ring denote multi-protein complexes. Connecting lines indicate interaction and arrows denote acting upon. Solid lines denote direct interaction while dashes are indirect interactions 139

158 factor for HIV-1 infection (Friedrich et al., 2011). Silencing of this protein led to a significant decrease in HIV-1 nuclear import. Thus, it is possible that this family of proteins could be important for HIV-1 replication and further studies will determine if there is a role for ADAM2 in viral replication. The last three network pathways identified involved proteins belonging to neurological disease (Fig4.9 A), cell morphology/cell development (Fig. 4.9B), and cellular assembly/development (Fig. 4.9C). These proteins involved in particular pathways will require further analysis to determine what role they may play in HIV-1 infection Validation of ProteoIQ data Next we attempted to validate several proteins identified in the infected samples by Western blot (Fig. 4.10). The RPS6 protein was used again as a control and was readily detected in all fractions (top panel). Next we probed the fractions for capping protein, gelsolin-like (CAPG) which functions in cell motility (Renz et al., 2008). CAPG was not detected in either the infected or uninfected samples (second panel), which suggested that it was falsely identified or Western blot was not sensitive enough to detect the protein. The next protein examined was the stem-loop binding protein (SLBP) which bind the stem-loop structure in histone mrna (Zhao et al., 2004). SLBP was detected in all fractions (third panel) despite it being identified only in the infected samples by MS. Lastly, the tuberous sclerosis 1 protein (TSC1) was queried in our proteomic fractions. Similar to CAPG, TSC1 was not detected in any of the tested fractions (bottom panel). Taken together, these results suggest that the data parsing technique used can greatly affect the final results obtained. For this study, we found the Proteome Discoverer software resulted in results that could be validated by other methods. However, the 140

159 141

160 Figure 4.9 The remaining pathways identified by ingenuity pathway analysis. (A) The neurological disease pathway. (B) The cell morphology/cell development pathway. (C) The cellular assembly/development pathway. Proteins unique to the infected samples are indicated in red and relative increased or decreased abundance is indicated by increasing intensity of red or green, respectfully. Circles containing an inner ring denote multi-protein complexes. Connecting lines indicate interaction and arrows denote acting upon. Solid lines denote direct interaction while dashes are indirect interactions. 142

161 Figure Validation of candidate proteins identified by ProteoIQ. Hypotonic gradient fractions were TCA precipitated, resuspended in sample buffer, and separated by SDS-PAGE. Several candidate proteins were detected using protein-specific antibodies. A whole C8166 cell lysate was used as a positive control to ensure proper detection of proteins. 143

162 ProteoIQ software, which uses a different database and different data sorting parameters, provided results that could not be validated. Overall this highlights the potential dilemmas that may occur when using different database searches in proteomic studies. 4.4 FUTURE STUDIES The proteomic screen of partially purified HIV-1 nucleoprotein complexes (NPCs) identified numerous cellular proteins which may play a role in HIV-1 infection. Indeed, this screen identified several known HIV-1 interacting proteins. To follow-up on this study, many of these proteins must be validated using the techniques employed in Chapter IV with LRPPRC. Knockdown by RNA interference will identify whether expression of a particular protein aids or hinders HIV-1 replication. To further confirm these results, co-localization using fluorescent microscopy will confirm association between HIV-1 and these cellular components. Studies mentioned above are currently ongoing to investigate several proteins identified above. This study also points to the potential problems of multiple database searching as the final data obtained varied greatly depending on the software used. Further analyses would need to be performed to validate the use of these data parsing software in proteomic studies. The problem may, in part, lie within the protein databases themselves since these are maintained by different groups and rely on experimental data for validation. 144

163 APPENDIX A: BARRIER TO AUTOINTEGRATION FACTOR AND RETROVIRAL INFECTIVITY. A.1. BACKGROUND The barrier to autointegration factor (BAF) is small 10 kda evolutionarily conserved protein. BAF is highly enriched in the nuclear envelope and interacts with several Lamina-associated polypeptide 2 (LAP2)-Emerin-MAN1 (LEM) domain proteins (Cai et al., 2001; Gruenbaum et al., 2002; Lee et al., 2001; Shumaker et al., 2001). BAF is thought to be involved in several nuclear functions including nuclear assembly and chromatin organization (Gorjanacz et al., 2007; Margalit et al., 2005), as well as gene expression (Dorner et al., 2006; Nili et al., 2001). BAF was originally identified by its ability to prevent autointegration of the murine leukemia virus (MLV) DNA (Lee and Craigie, 1998) and later isolated from preintegration complexes from cells infected from MLV and HIV-1 (Lin and Engelman, 2003). Removal of BAF from MLV PICs caused self-destruction of the virus by autointegration (Suzuki and Craigie, 2002). However, knockdown of BAF in permissive cells did not affect HIV-1 or MLV infection (Shun et al., 2007a). Additional evidence suggested that BAF may interact directly with HIV-1 proteins Pr55 Gag and matrix (Mansharamani et al., 2003), but a recent study failed to detect direct interaction with HIV-1 matrix in the absence of DNA (Huang et al., 2011). Thus, interactions between BAF and PICs are likely due to non-specific DNA binding. Evidence to support this idea includes BAF relocalizing and condensing transfected DNA from several sources (Ibrahim et al., 2011). This study also demonstrated potent antiviral function for BAF in vaccinia virus infected cells by viral DNA condensation. Vaccinia virus overcomes this repression by specifically phosphorylating BAF to relieve 145

164 DNA binding. Thus, it is possible BAF may potentially inhibit other viruses such as HIV- 1. In this study we examined HIV-1 infectivity in several BAF expressing cell lines to determine corroborate the role of BAF in HIV-1 infection. We found that overexpression or knockdown of BAF had no effect on HIV or MLV infectivity. Additionally, there was no co-localization observed between HIV-1 and BAF during early infection. Taken together, these data confirm that BAF is not a critical factor for HIV-1 infection. A.2 MATERIALS AND METHODS A.2.1 Cell Culture and Virus The CV-1 shns and shbaf cells were obtained from Dr. Matthew Wiebe and were maintained in complete DMEM-10%-FC3 supplemented with 1 µg/ml puromycin (Ibrahim et al., 2011). The 293T inducible cells (WT and MAAAQ) were also provided by Dr. Matthew Wiebe and were maintained in DMEM-10%-FC3 supplemented with 100 µg/ml hygromycin. All MLV-Luc and NLX-Luc virus was produced as described in section The NLX-Cherry-Vpr virus was produced by transient transfection of 293TK cells with 15 µg of pnlx-cherry-vpr and 5 µg of pmd2.g. Virus was harvested and concentrated as outlined in section A.2.2 Luciferase Assays The luciferase assays were performed essentially as described in Section The inducible cells were induced 24 hours prior to NLX-Luc transduction with 0.1 µg/ml doxycycline. Overexpression of Flag-BAF protein was monitored by Western Blot using an anti-flag antibody. 146

165 A.2.3 Fluorescent Microscopy Wild-type Flag-BAF and MAAAQ Flag-BAF cells were seeded on glass coverslips. Cells were induced by 0.1 µg/ml doxycycline 24 hours prior to infection with Cherry-Vpr NLX + VSVg. Virus was removed after 4 hours and the cells incubated an additional 24 hours. The cells were washed with PBS, fixed with 3.7% formaldehyde, and permeabilized with 0.2% Triton X-100. The cells were incubated with anti-flag monoclonal antibody for 1 hour, cells were washed thoroughly with PBS, and the secondary anti-mouse FITC antibody added for 1 hour. Cells were washed and mounted on glass slides using ProLong anti-fade reagent (Life Technologies, Grand Island, NY). Slides were examined and images were captured using a Nikon Eclipse 80i microscope (Nikon Instruments, Melville, NY). A.3 RESULTS Previous results demonstrate association of BAF with HIV-1 PICs, but knockdown or knockout of BAF did not significantly inhibit HIV-1 infection. Conversely studies showed MLV infection required BAF to prevent autointegration, but knockdown of BAF did not significantly decrease MLV infectivity. This conflicting data suggests that BAF-PIC association may not be biologically relevant. To test if BAF expression was required for retroviral infectivity we transduced a previously described 293T cell line stably expressing a potent shrna against BAF with MLV-Luc (Figure A.1). We observed no change in transduction efficiency of MLV-Luc in shbaf cells compared to the control shns cells. This result suggests that BAF is dispensable for efficient MLV infection. 147

166 Figure A.1. Knockdown of BAF does not decrease MLV infectivity. Stable shns and shbaf CV-1 cells were transduced with MLV-Luc + VSVg. Cells were lysed 48 hpi and luciferase activity was measured. Data is representative of 3 independent experiments and error bars denote SEM. 148

167 The DNA binding properties of BAF are controlled through phosphorylation (Bengtsson and Wilson, 2006; Nichols et al., 2006). Unphosphorylated BAF will not specifically bind DNA, while phosphorylation by several vaccinia-related kinases (VRKs) relieves DNA binding. Recent work demonstrated that a BAF mutant (MAAAQ) is unable to become phosphorylated by cellular VRKs and irreversibly binds DNA from several sources (Ibrahim et al., 2011). This mutant also has a potent antiviral effect against vaccinia virus in vivo. Thus, we tested the ability of this MAAAQ mutant to inhibit HIV-1 infection. Inducible 293T WT and mutant BAF cell lines were induced with doxycycline 24 hours prior to infection with NLX-Luc. HIV-1 transduction was not altered in both wild-type and MAAAQ BAF induced cells compared to the respective uninduced cells (Figure A.2). The Western blot below the graph demonstrates expression of both BAF proteins when doxycycline is present. These data suggest that mutant MAAAQ BAF protein has no antiviral effect on HIV-1. Lastly, colocalization of BAF with HIV-1 was examined during early infection. The inducible Flag-BAF cell lines were induced by doxycycline and infected with a HIV-1 clone harboring a Cherry-Vpr (NLX+Cherry-Vpr). The cells were incubated with anti- Flag-FITC antibodies, washed, and fixed. Fluorescent microscopy revealed that BAF was mostly relegated to the nucleus in these cells (Figure A.3, both panels green). After infection with NLX-Cherry-Vpr we observed no co-localization between Flag-BAF and Cherry-Vpr (Figure A.3 bottom panel, white arrows). This suggested that BAF does not transport to the cytoplasm during HIV-1 infection and is not found in viral complexes containing Vpr. 149

168 Figure A.2. WT and mutant MAAAQ BAF does not alter HIV-1 infectivity. Inducible BAF cell lines were cultured with (+) and without (-) doxycycline and 24 h later transduced with NLX-Luc + VSVg. Data is normalized to the non-induced WT BAF cells. Western blot for Flag-BAF and actin is shown below graph. Data is representative of 3 independent experiments and error bars denote SEM. 150

169 Figure A.3. BAF does not colocalize with HIV-1 during early infection. Flag-BAF inducible cell lines were infected with NLX + Cherry-Vpr virus 24 hours post induction. Flag-BAF was detected using anti-flag antibody conjugated to FITC. Arrows denote infected cells as indicated by Cherry-Vpr signal. 151